Migration in Africa: A Review of the Economic Literature on
International Migration in 10 Countries
William Shaw
Development Prospects Group
The World Bank
Washington DC 20433
April 2007
Introduction
This paper reviews the economic literature on international migration in 10 African countries
(Burkina Faso, Ethiopia, Ghana, Lesotho, Mali, Mauritius, Nigeria, Senegal, South Africa and
Uganda), as a contribution to a larger study on migration in Africa.1 While the selection of
countries is motivated by the plan of the larger study, these countries serve as useful illustrations
of the range of migration experiences in Africa. Included are some of the poorer and some of the
(relatively) richer countries in the Sub-Saharan Africa region; stable countries and others that
have suffered internal and external conflicts; and countries that serve as destinations for migrants
from other African countries but are sources of migrants to Europe and America, others that are
both source and destination countries for African migrants, and some that have shifted from
principally destination to source countries over time. This review will focus on the economic
implications of migration, while recognizing that migration also may have important social and
political impacts.
We will begin with a summary of the conclusions. The main body of the paper covers an
overview of the available data on migration in the 10 countries, the benefits and risks involved in
migration, the impact of migration on destination countries (an area where lack of data seriously
constrains analysis), the implications for origin countries (remittances, other diaspora benefits,
and the brain drain), and a discussion of health aspects of migration. We will conclude with a
discussion of priorities for future research, from the perspective of next steps in the broader
Africa research project. The paper attempts to provide an integrated view of the migration
experience in the 10 countries based on examples from the literature, rather than a
comprehensive survey of everything that has been written. The annex covers additional material
and data on each country that do not fit well into the main study.
Main conclusions
Given the scarcity of information, most statements on migration in Africa need to be
heavily qualified. The following are reasonable conclusions that emerge from the economic
literature on the 10 countries:
(i) Most of the migration from (and to) the sample countries is from neighboring
countries, and seasonal migration to take advantages of differences in growing seasons plays an
important role. Some countries do show a significant share of their total emigration going to
Europe and North America, although to some extent this may reflect the lack of data on
migration within Africa.
(ii) Economic motivations for migration include to improve earnings (either immediately
or over the long term through training) and to diversify earnings sources to reduce the risks
facing households. Migration requires resources, so that emigrants to Europe or North America
tend to be wealthier and more educated than emigrants to other African countries, while often the
1
A project on African migration, funded by the World Bank, the African Development Bank, and UK’s DFID, will
undertake literature reviews, surveys, and interviews with policy makers. Dissemination will be through published
materials and seminars in Africa. The author is grateful to the African Development Bank for funding to write this
paper, and to Dilip Ratha for his leadership in initiating the project.
2
poorest cannot migrate. Nevertheless, studies indicate that migration plays an important role in
reducing poverty, mainly through remittances, and also contributes to smoothing consumption of
the poor. Networks play an important role in reducing the costs and risks of migration, and may
explain why individual regions or communities account for a large proportion of emigrants from
some of the countries.
(iii) The sample countries impose few restrictions on emigration, while legal limits on
immigration have proven difficult to enforce. Expulsions of immigrants have reduced the stock
of foreigners (at least for a few years), but at terrible costs for the migrants. Undocumented
immigration is currently a subject of great controversy in South Africa, where large differences
in income with neighboring countries creates substantial pressures for immigration. Mauritius,
due to its geographical isolation, is the only one of the 10 countries that has been capable of
effectively controlling irregular immigration. Only limited efforts have been made to promote
the welfare of emigrants and to ensure the free movement of people within regional
organizations.
(iv) Exploitation and abuse of women and children are all too frequent in most of the
sample countries. Frequent migration by children and child fostering in West Africa increase the
vulnerability of children and make it difficult to identify the victims of abuse. A range of
coercive and misleading practices are used in the trafficking of women, which can be
compounded by rules in destination countries that leave women in the power of employers. In
several of the sample countries, inadequate legislation, poor enforcement, social attitudes and the
legality of prostitution make trafficking difficult to combat.
(v) Remittances to the sample countries have increased sharply over the past decade in
dollar terms and in relation to GDP. However, official data likely underestimate the size of
remittances, as several observers report very large shares of remittances going through informal
channels. Econometric analyses have shown that remittances contribute significantly to reducing
poverty in Burkina Faso, Ghana, and Lesotho. In countries and regions with massive emigration,
there is some evidence that the receipt of remittances may reduce work effort.
(vi) Means of transferring remittances used in the sample countries include banks or
other financial intermediaries, post offices, money transfer operators (e.g. Western Union and
Moneygram), carrying cash by hand (personally or through an agent), bringing goods, instructing
payment through cell phones or other communications media (with settlement within the transfer
firm effected separately), bundling remittances through informal networks to reduce transactions
costs, and purchasing goods or services for the benefit of households in origin countries. Fees in
informal networks tend to be lower than in banks or money transfer operators, although not
necessarily lower than postal orders (except post offices can be slower and less reliable than
other transfer modalities). In some countries, regulatory restrictions impede entry into the
remittance market of financial institutions with extensive geographical reach and proximity to
the poor.
(vii) It is extremely difficult to measure the impact of skilled emigration on origin
countries. Several of the sample countries have a large share of their college-educated people
living overseas. But in many countries a poor investment climate reduces the demand for skilled
3
professionals, while educational policies contribute to a sector-specific surpluses and deficits of
highly-educated workers by producing more social science graduates than engineers and
scientists. Data on the emigration of health professionals, coupled with shortages in public health
employment, are quite alarming. Yet in some countries funding is insufficient to hire the
available nurses registered as unemployed.
(viii) Governments have tried to encourage return (either temporary or permanent) of
skilled emigrants through subsidies, provision of information, cultural exchanges, allowance of
dual citizenship, and maintenance of contacts over the Internet. However, subsidy programs
(including those financed by destination countries) have been limited and have had mixed
results, while the impact of other efforts is unclear. Fragmentary survey data indicate that
returned migrants may have improved their productivity while abroad.
(ix) Migration can encourage the spread of communicable disease as people carry
viruses over long distances, and migrant populations tend to be vulnerable due to lack of health
services and, in the case of AIDS, greater likelihood of risky sexual behavior.
Migration trends
Migration has a long tradition in Africa, for example from the 4th century in Mali
(Findley 2004). Emigration has had an important impact on all of the sample countries, and in
several immigration also has played an important role. The sample countries show a wide
variety of international migration patterns. Some migrants pursue seasonal agricultural activities
that involve crossing borders; this form of migration may become a regular pattern of
employment over long periods of time. Others migrate for short periods of time as a reaction to
changes in economic conditions, as a means of gaining experience or training, or to save for a
future investment. Individuals or whole families may choose to settle in other African countries
to achieve a better way of life. Or people may migrate to Europe, America, or Asia on a shortterm or long-term basis. Crossing borders for the purpose of trading is common in Africa, and
depending on the length of stay may not be thought of as migration. Travel for cross-border
trading also occurs between Africa and Europe. For example, Senegalese traders in Italy return
on a regular basis to Senegal to re-stock their merchandise (Riccio 2003).
The data on migration in developing countries is poor, due to weaknesses in agencies
responsible for collecting data and the lack, or easy avoidance, of border controls. These
weaknesses are not unique to Sub-Saharan Africa (note the example of undocumented
immigration to the United States), and their severity varies considerably among the sample
countries.
We will first consider the 10 countries from the perspective of destinations for migration,
and then as sources of migrants. This distinction is a useful organizing principle, but has some
drawbacks. Some countries are at the same time destination, origin, and transit countries.
Others, including Ghana, Senegal, and Uganda have shifted over time from important
destinations for migration to principally sources of emigrants with the deterioration in economic
conditions.
4
Viewing the sample countries as destinations for immigration
According to official data, in most of the sample countries the share of immigrants is less
than 3 percent of the population (the exceptions are Ghana with 7 percent and Burkina Faso with
just under 6 percent). The number of foreigners in the sample countries fell from 2.9 percent of
population in 1960 to 1.7 percent in 2005 (table 1). This trend reflects the rapid population
growth in most of these countries (averaging 2.6 percent per year for the group as a whole), and a
decline in attractiveness of migration due to conflict and economic crisis. Uganda, for example,
was a relatively prosperous and stable country in the 1960s, when the stock of immigrants
equaled more than 11 percent of the population. But with the economic and political decline in
the 1970s and 1980s, Uganda became an important source of emigrants (Black and others 2004),
and the total number of immigrants has fallen by one third, to 1.8 percent of the population.
During the relative prosperity of the early 1960s, Ghana was an important destination country for
migrants from neighboring West African countries, but immigration fell with the economic crisis
after 1965, and dropped further by the 1980s with shortages of basic consumer goods (Anarfi and
others 2003). Immigration has recovered strongly with the return of stability and economic
growth, and in 2005 the stock of immigrants had returned to about the level of 1960 as a share of
population. The stock of immigrants fell in Nigeria in the mid-1980s due to the expulsion of
foreigners, and in South Africa through the 1990s with the drop in demand for mine workers
(although here the data may be distorted, as undocumented flows probably increased after
apartheid ended and the anti-immigrant atmosphere would not encourage honesty in filling out
census forms). Senegal also shifted from mainly a country of immigration in the 1970s to one of
emigration in the 1990s (Fall 2003).
Table 1: Stock of immigrants as a share of population
(percent)
Burkina Faso
Ethiopia
Ghana
Lesotho
Mali
Mauritius
Nigeria
Senegal
South Africa
Uganda
Total
1960
1.4
1.7
7.7
0.4
2.9
1.6
0.2
4.8
5.4
11.7
2.9
1965
1.6
1.5
5.3
0.3
2.3
1.4
0.3
4.4
4.8
11.6
2.6
1970
1.7
1.3
3.9
0.3
1.7
1.3
0.3
3.8
4.2
10.3
2.3
1975
1.8
1.1
3.8
0.3
1.3
1.1
1.2
2.9
3.7
7.2
2.2
1980
2.6
1.1
3.7
0.7
1.0
1.0
1.9
2.0
3.4
5.4
2.3
1985
3.7
1.3
3.7
1.1
0.8
0.9
0.4
2.5
5.5
4.3
2.1
1990
4.0
2.3
4.6
0.4
0.7
0.8
0.5
3.7
3.3
3.1
2.0
1995
4.7
1.3
5.9
0.3
0.6
1.0
0.6
3.5
2.6
2.9
1.8
2000
5.1
1.0
7.6
0.3
0.4
1.3
0.6
2.9
2.2
2.2
1.7
5
2005
5.8
0.7
7.5
0.3
0.3
1.7
0.7
2.8
2.3
1.8
1.7
Figure 1: Stock of immigrants by number and percent of population
Sources: UN Population Division and DEC Prospects Group, the World Bank
Other countries have experienced relatively stable numbers of immigrants in relation to
population. The number of immigrants is minimal in Lesotho, Mali, and Mauritius, the latter
because of its geographical isolation that facilitates effective barriers against irregular
immigration, the first two because of the depth of poverty and lack of employment opportunities.
Where data are available on source countries, almost all immigrants in the sample
countries are from Africa (figure 2), and largely from neighboring countries. Most of the
identified migrants in Burkina Faso are from neighboring countries, about a third of these
victims of forced migration. Almost 60 percent of Ethiopia’s immigrants are from Eritrea after
the war separating the two countries, and another 30 percent are from (even poorer) Somalia.
The bulk of identified immigrants to Senegal come from neighboring Guinea, Mali, and
Mauritania. And about 80 percent of Uganda’s immigrants come from nearby countries that have
suffered from sustained violence—Burundi, Democratic Republic of the Congo, Rwanda, and
Sudan. Nigeria, despite its severe economic difficulties, continues to attract immigrants from
other West African countries. South Africa is the only middle-income country in the sample
with land borders, and attracts very significant inflows of people seeking work. The low share of
immigrants from Africa in Burkina Faso reflects a large number of immigrants for whom no
source country is given in the data, and in Lesotho reflects a recent influx of Chinese immigrants
in conjunction with the opening of textile factories financed by Taiwanese investors (Cobbe
2004).
6
Figure 2: Share of immigrants from Africa
(percent of total immigrants, 2005)
Uganda
South Africa
Senegal
Nigeria
Mauritius
Lesotho
Ethiopia
Burkina Faso
0
20
40
60
80
100
120
Note: No official data are available on the source countries for immigrants to Ghana
Source: DEC Prospects Group, the World Bank
Unlike the trend for developing countries as a group, female immigration has risen
sharply in the sample countries, from 27 percent of total immigrants in 1960 to 49 percent in
2005. It is not clear how reliable these data are, as three countries (Lesotho, Mali and Senegal)
report exactly the same ratio of females to total immigration in every reported year from 19602005. Nevertheless, the three countries with the largest number of immigrants (Ghana, Nigeria
and South Africa) all show sharp increases in the share of female migration. It is possible that
these data reflect increasing participation by women in economic activities outside the home. It
is also possible that increased trafficking of women makes a contribution to this increase,
although in the absence of data it is difficult to say whether the numbers are large enough to
affect the overall share of females in international migration.
Viewing the sample countries as sources for emigrants
Considerable efforts have been made to estimate the number of emigrants from
developing countries, and where they have emigrated to, based on reports of the allocation of
immigrants in partner countries. The data presented here are taken from the World Bank’s
Development Prospects Group.2 Given the shortcomings of the partner country immigration
data, there inevitably are considerable gaps. Emigration is relatively high in Burkina Faso and
Mali (largely to each other and Cote d’Ivoire), Lesotho (to South Africa), and Mauritius (where
higher incomes facilitate emigration, largely to Europe and Australia). Emigration from Ghana
(more than a third to Cote d’Ivoire) and Senegal (with very little information on the breakdown
within Africa) equal 4 percent of the population. Emigration from the rest of the sample
countries is below 2 percent of the population, according to these estimates.
2
The raw data can be found at www.worldbank.org/migrationandremittances. Click on ‘Development
Prospects Group’ in the right-hand box, and then on ‘migration data’.
7
Figure 3: Emigrants by percent of population and number of people
Total Emigration, 2005
Emigration by Destination Region, 2005
(percent of population)
(number of people)
Uganda
Uganda
South Africa
South Africa
Senegal
Senegal
Nigeria
Nigeria
Mauritius
Africa
Other Developing
High Income
Mauritius
Mali
Mali
Lesotho
Lesotho
Ghana
Ghana
Ethiopia
Ethiopia
Burkina Faso
Burkina Faso
0
2
4
6
8
10
12
14
16
0
500,000
1,000,000 1,500,000
Source: DEC Prospects Group, the World Bank
Most emigrants go to other African countries, and indeed neighboring countries. Much
of this may be seasonal migration, either pastoralists leading their animals across borders for
better grazing, or more commonly travel to take advantage of differences in growing seasons. In
West Africa, the main international migration movements are to supply labor for commercial
agriculture from Burkina Faso to Cote d’Ivoire and Ghana, and from Mali and Guinea to Cote
d’Ivoire and Senegal (Adepoju 1988). Many households in Mali send a large share of their
active labor force to work on coca farms in Cote d’Ivoire, but return home to cultivate cotton and
grain during the short Sahelian rainy season (Konseiga 2005b).3 World Bank (1993) finds that
most migration is seasonal in Mali. In Burkina Faso, the estimated share of total emigrants in
neighboring countries is 90 percent, and in Mali it is 87 percent. In Ethiopia, while long-term
migration is seen by most poor families as an option of last resort, temporary seasonal migration
is often encouraged as a rite of passage for young men (Black and others 2004). In Ghana,
regional variations in the seasonality of agricultural production promote migration.
Surprisingly, according to these estimates emigration to high income countries
constitutes more than half of total emigration for Ethiopia, Mauritius, Nigeria, South Africa, and
Uganda. For Mauritius and South Africa, incomes are high enough that emigrants are typically
attracted to the even higher-income countries. Emigration to Europe is significant for Nigeria.
But nevertheless one suspects that for that country, as well as Ethiopia and Uganda, the relatively
high share of high-income countries in emigration reflects massive underestimates of emigration
to neighboring countries.
Determinants of migration
Here we will consider economic determinants of migration, including income
differentials between origin and destination countries, the desire to reduce risks by diversifying
3
In part this practice results from the extreme labor intensive nature of farming in Mali, with little use of
animals or mechanical aids. If families cannot hire someone to replace a migrant, then the migrant must
return to help with cultivation during the rainy season. (Findley and Sow 1998)
8
the sources of income, threshold effects, forced migration, and the influence of networks. Social
structures and attitudes that condition migration, while important, are beyond the scope of this
inquiry.
Income differentials
While economists may assume that migration (leaving aside forced migration) is
generally an economic phenomenon, it can be difficult to establish this empirically. Myburgh
(2004) finds a positive correlation between emigration from South Africa to the United States,
the United Kingdom, Australia and New Zealand and the gap in average annual wages between
origin and destination countries, from 1987-99. Nevertheless, barriers to immigration make it
difficult to estimate the relationship between income differentials and the (unconstrained) supply
of emigrants.4 Some writers get at this issue through surveys. In Nigeria, for example, a survey
found that reasons for migration were predominantly economic (80 percent of respondents) or
for education (16 percent) (Nwajiuba 2005). But Guilmoto (1998), in a study of the Senegal
River valley, emphasizes the influence of social mores in conditioning migration, particularly as
migrants are heavily dependent on the family or village for advance money and access to
networks. Efforts to measure the income benefits from migration are relatively rare, as they
would require considerable data on earnings in source and destination countries and on migrants’
qualifications.
Diversification
Migration also can be motivated by the desire to diversify the risks facing the poor, by
placing household members in distant countries or regions where adverse shocks are not
correlated with those facing the home region (Lucas and Stark 1985). While much of the data
work supporting this view is based on internal migration, in Africa borders are sufficiently
porous that motivations for internal and international migration are often similar. Migration may
also compensate for imperfect credit markets that make it difficult or impossible for individuals
to borrow in order to invest. Survey data collected in northeastern Burkina Faso show that
seasonal migration enables households to invest in their main pastoral activities (Konseiga
2005b).
Threshold effects
Migration, and in particular international migration, often is not within the ability of the
poorest. This is most evident with overseas migration. For example, a survey of rural households
in four villages in Burkina Faso found that intercontinental migrants tended to come from the
highest-income group and to earn more in remittances than continental migrants, while
households with intercontinental migrants have higher income and are better off in terms of land
and livestock than households with continental migrants (Black and others 2005). Wouterse and
van den Berg (2004) found that migration to other African countries from Burkina Faso was
undertaken by comparatively poor households in response to lack of work and insufficient
income, while migration to Europe was by more wealthy households in response to opportunities
for accumulation of wealth. Similarly, Findley (1989) finds that emigrants to France from the
4
A few articles have used distance as a proxy for cost, and found a significant relationship between distance and
migration flows between countries (see, for example, Adams and Page 2003).
9
Senegal River valley are most likely to come from villages with irrigation facilities installed
before 1978, the beginning of the recent wave of migration.5 But even continental migration
may be subject to threshold effects. A study of the Fulani, an ethnic group in northern Burkina
Faso, found that seasonal labor participation increased with higher household wealth, as low
financial and human resources limit the ability of the very poor to migrate (Hampshire and
Randall 1999). Black and others (2004) found that among poor Ethiopians, those with some
assets, rather than none, were more likely to migrate.
Education levels also are frequently found to be tied to the ability to migrate. Wouterse
and van den Berg (2004) also found that the greater number of persons in a household with
secondary education, the greater probability that someone from the household would migrate to
another African country (controlling for income levels), perhaps due to greater access to
information. Survey data show that migrants from Ghana, particularly international migrants,
are more likely to be highly-qualified than non-migrants (Lichfield and Waddington 2003), and
that the probability of migration rose with education (Tsegai and Plotnikova 2004). By contrast,
Findley (1989) finds that there is no evidence that migration rates from the Senegal River valley
are higher for literate individuals or communities with good educational facilities, and speculates
that literacy levels are generally too low to distinguish the effect (80 percent of the sample is
illiterate in French).
Forced migration
Refugee flows have accounted for a small share of migration both from and to the sample
countries. Uganda had the largest number of victims of forced migration in 2005, amounting to
250,000 people or equal to almost half the numbers of immigrants in the official data, followed
by South Africa and Ethiopia (figure 4).6 Ethiopia was the source of the largest number of
forced migrants. In line with a decline in violence and efforts at resettlement, the number of
forced migrants in the sample countries declined from over a million in 1995 to 647 thousand
persons in 2005.
Figure 4: Forced migration
Forced Migration by Destination, 2005
(number of emigrants)
Forced Migration by Origin, 2005
(number of emigrants)
300,000
250,000
200,000
150,000
100,000
50,000
Burkina Faso
Ethiopia
Ghana
Lesotho
Mali
Mauritius
Nigeria
Senegal
South Africa
Uganda
0
Includes refugees, asylum seekers, displaced persons, and stateless persons.
Source: United Nations High Commission on Refugees
90000
80000
70000
60000
50000
40000
30000
20000
10000
Burkina Faso
Ethiopia
Ghana
Lesotho
Mali
Mauritius
Nigeria
Senegal
South Africa
Uganda
0
Includes refugees, asylum seekers, displaced persons, and stateless persons.
Source: United Nations High Commission on Refugees
5
Quoted in Russell and others (1990).
Here we are defining forced migrants as the UN High Commission on Refugees’ expression “persons of concern”.
This includes some internally displaced persons.
6
10
We are not here concerned with the determinants of forced migration, efforts to assist
forced migrants, or resettlement. However, it is useful to note one issue in the literature about
the potential for economic integration of refugess. A study of self-settled refugees in Uganda
suggests that many of them have been able to find employment and have either achieved selfsufficiency or are on the way to doing so (Macchiavello 2003). After the flight from Cote
d’Ivoire by citizens of Burkina Faso in the mid-1990s (see below), the government decided to
encourage the reintegration of returned migrants into their original communities (although thirdor fourth-generation returnees who had been born in Cote d’Ivoire and had limited connection to
any village generally stayed near the frontier—Pizarro 2006). Many were taken in by relatives,
and as a result there was no permanent burden placed on the government to support refugees. On
the other hand, the lack of refugee camps meant that little international attention was paid to the
returnees, so aid resources were limited.7 South Africa has had some success in integrating longterm refugees from Mozambique, including efforts to provide citizenship to some (GoloobaMutebi and Tollman 2004). These experiences are consistent with those in other countries where
efforts were made to integrate refugees.8
Influence of networks
Networks can provide information on migration opportunities, facilitate finding
employment, and support migrants while looking for work, thus lowering the costs and risks of
migration. There is considerable evidence of the influence of networks on migration flows.
Previous colonial bonds continue to have impact on migration flows, in part due to common
language and well-established networks (Hearing and van der Erf 2001). In a 1995 survey, 23.5
percent of Ghanaian migrants stated that they emigrated because of the presence of relatives and
friends (Anarfi and others 2003). Ghanaian diaspora associations, such as Pentecostal churches,
assist new migrants in the United States, the United Kingdom, Canada and other countries
(Quartey 2006). Afolayan (2001) notes that social networks that have evolved over the years
have led to the concentration of Malian emigrants in certain destinations in France. The
importance of networks may also explain the concentration of emigrants from relatively limited
areas of origin countries. For example, a disproportionate share of Senegalese emigrants come
from the regions of Saint-Louis and Matam in the Senegal river valley, as well as the cities of
Dakar and Touba (Sander and Barro 2003); the Ashanti and greater Accra region account for
most international remittances to Ghana (Kabki and others 2004, Mazzucato and others 2005);
and the Kayes region of Mali claims a disproportionate share of Malian émigrés.
Networks may encourage emigration in part by demonstrating its benefits and feasibility.
In a study of the Senegalese diaspora, Ricci (2003) shows how the success of trading networks
between Senegal and Europe encouraged more emigration. In addition, existing trading
networks provided market information and credit to newcomers. Networks may also act to
perpetuate migration even after the initial causes have waned. Emigration flows from Ghana
were initially spurred by the economic crisis, but appear to continue to increase despite that
7
Donors did finance first aid, reception centers, food and emergency relief (particularly vaccination of children) for
returnees at the border, but no permanent assistance (Pizarro 2006)
8
Banki (2004) discusses the varying experience of Tibetan and Bhutanese refugees in Nepal. Tibetan refugees were
able to find productive employment in neighboring villages, as the Nepalese authorities were reluctant to maintain
large refugee camps along the border with China. By contrast, refugees from Bhutan, who culturally are much
closer to the Nepalese, are maintained in camps with little opportunity to become self sufficient.
11
country’s recovery, due to the establishment of networks (Higazi 2005). Note, however, that at
the same time immigration to Ghana has recovered sharply, according to the official data.
Guilmoto (1998) notes that emigration fell only slowly after the introduction of irrigation in the
Senegal river valley, likely due to the momentum from networks and the population becoming
accustomed to emigration. Bach (2003) observes that overseas nurse associations and other
networks in the United Kingdom have supported further emigration of nurses from South Africa.
Government management of migration
The freedom to emigrate is recognized in some constitutions (for example in Burkina
Faso—Ba 2005), and in general restrictions on emigration are minor (Russell and others 1990).
But immigration policy is more problematic.
Restrictions on immigration
Migration into many of the sample countries is generally unencumbered by official
restrictions, either as a matter of policy or in practical terms, given weak government abilities to
control immigration. For example, most migrants enter or leave Burkina Faso without proper
papers (Pizarro 2006). Even after independence, migration between Ghana and Togo, as well as
Nigeria and Benin, was undocumented and unhindered by long, unpoliced borders lacking
physical landmarks (workers often commuted daily across the border) (Adepoju 2005). The 1995
constitution in Uganda provides for the establishment of a national Citizenship and Immigration
Board, but several years later researchers could find no evidence of its existence (Black and
others 2004). Nevertheless, there are important exceptions to this generally permissive
environment. Some countries have received immigrants for decades, but then shifted to mass
expulsions (e.g. Nigeria, Uganda) during economic difficulties (see discussion under risks to
migrants). Mauritius has little trouble controlling immigration, with the exception of a few cases
of visa overstays. Mauritius has specific programs to divide legal entrants between less-skilled
workers that are eligible only for temporary work permits and more skilled workers and investors
for whom permanent residence is possible (Hein 2004).
South Africa does attempt to control irregular migration, but with limited success.
Border posts are understaffed and lack basic facilities such as facsimile machines, consistent
electric power supply, proper living quarters for officers, vehicles, and proper search and storage
facilities. Control also is hampered by rampant corruption and a lack of trust between the
agencies involved (Hennop and others 2001). The annual number of visitors from countries of
the Southern African Development Community rose from 1 million in the early 1990s to over 5
million a few years ago, overwhelming the ability of border posts to cope and leading to delays,
inefficiencies, and corruption as people tried to jump the queue (Black 2004). Estimates of the
number of irregular migrants in South Africa vary widely (Maharaj 2004), with some observers
claiming figures as high as 8 million (Solomon 1996). However, the formulae used to compute
such estimates are essentially extrapolations from the number of repatriations from South Africa
and overstays, and are extremely unreliable. In a country with a population of 47 million, such
divergent views of the magnitude of irregular migration make it difficult to estimate the impact
of immigration. Some writers are primarily concerned with protecting irregular migrants from
abuse by authorities and employers. For example, Klaaren and Ramji (2000) cite widespread
12
abuses of immigrants’ rights and corruption as an institutional feature of arrest and detention of
undocumented migrants, while Crush (2003) claims that methods of arrest and removal of
undocumented migrants haven’t changed since the apartheid era. Others decry the burden on
public services and infrastructure, the contribution to crime (particularly the smuggling of drugs
and weapons—Minnaar 2001), and the competition facing low-wage workers (see below).
International agreements
Some of the sample countries have made efforts to promote their emigrants or improve
their conditions through bilateral agreements. In the 1960s, exploitation in destination countries
led Upper Volta to sign agreements with Ivory Coast and Gabon to humanize working conditions
and better monitor migrant flows. The agreements provided for the social protection of migrants,
and the creation of recruitment and transit centers financed by the destination countries.
However, the results were disappointing, as Burkinabe refused to migrate within the official
framework, and the main clauses were not respected by Ivoirian farmers (Konseiga 2005a).
Senegal signed an agreement with Gabon in 1987 securing social security benefits for children of
Senegalese workers (Diatta and Mbow 1999). More recently, Senegal signed an agreement with
France providing that families of Senegalese workers in France could receive benefits under
French social safety net provisions. As a result, Senegalese families with members in France
receive payments under Caisse des Allocations Familiales. (Ammarrasi 2005).
Provisions for the freer movement of persons have also been pursued within regional
agreements. The promotion of the free movement of persons within the union is a long-term
objective of the Economic Community of West African States (ECOWAS), the Southern African
Development Community (SADC), and the Common Market for Eastern and Southern Africa
(COMESA). However, only ECOWAS, which comprises Benin, Burkina Faso, Cape Verde,
Côte d'Ivoire, The Gambia, Ghana, Guinea, Guinea Bissau, Liberia, Mali, Niger, Nigeria,
Senegal, Sierra Leone, and Togo, has actually implemented steps to promote free migration.
Attempts were made in the 1980s to implement protocols on the free movements of citizens, but
these did not stop Nigeria from expelling aliens (mostly Ghanaians) in 1983 and 1985, and Cote
d’Ivoire from doing the same in 1986. Efforts continued, however, culminating in an agreement
to permit the entry of citizens from other members for 90 days without visa requirements
(Adepoju 2002). A new treaty in 2003 recognized the right of free movement of community
citizens and the right to work in member countries, although obstacles to implementation of the
regulations emerged in Cote d’Ivoire (Konseiga 2005a) and as of last year only the entitlement to
travel without visa for up to 90 days had been implemented (Pizarro 2006). Citizens of the East
African Community and the Common Market for East and Southern Africa now enjoy visa-free
entry in member countries (Oucho 2006), which has reduced administrative burdens facing some
migrants. A SADC protocol that would provide for visa-free entry up to 90 days per year of
nationals from other member states was adopted at the summit in August 2005, and by October
had been signed by half of SADC members, with two-thirds required for ratification
(Madakufamba 2005). However, as of February 2007 it appears that exemptions from visa
requirements for citizens traveling between southern African states will not be implemented until
2008 (SADC Today 2007). Overall, the African regional organizations have taken steps to
facilitate short-term stays in member countries, but the establishment of large economic unions
within which citizens could move and work freely remains a longer-term goal.
13
Several African countries also participate in consultation processes to facilitate regional
cooperation in managing international migration, including the Migration Dialogue for Southern
Africa, and conferences to discuss migration in Eastern Africa (Ammarrasi 2005), as well as the
Dakar Declaration initiated by the International Organization for Migration in 2000 and signed
by most West African countries (Black 2004). It does not appear that evaluations of the impact
of these regional processes are available.
Risks to migrants
While migration generates enormous benefits, it also can subject migrants to abuse. In
the experience of the sample countries, governments have imposed suffering on migrants
through mass expulsions, while private individuals have used deception and violence to condemn
women and children to exploitation.
Mass expulsions
Violence and governmental decisions have imposed severe hardships on migrants. In
Cote d’Ivoire, deteriorating economic conditions in the 1990s, measures to disenfranchise
migrants and deport those without proper documents, and the wave of violence against foreigners
that began early in this decade, led to more than 365,000 Burkinabe fleeing home (IOM 2005b,
Antropologi 2005). Nigeria expelled about 2 million foreigners with the decline in economic
conditions in 1983 (Lassailly-Jacob and others 2006), compounded by beliefs that foreigners
were implicated in the social unrest in 1982 and a desire to minimize security problems leading
up to the 1983 elections (Afolayan 1988). Thousands of Ghanaians died on the way home, and
returnees suffered from a critical shortage of food (Higazi 2005).
Exploitation/abuse and trafficking
Undocumented migrants tend to be vulnerable to abuse. In South Africa, sectoral studies
reveal widespread economic and sexual exploitation in sectors with large number of irregular
migrants, accompanied by substantial fear among migrants (Black 2004). Trafficked women can
even find difficulties in obtaining public help. For example, in South Africa many shelters for
battered women require a South African identity card as a condition of entrance.
Extent. Illegal activities are by their nature difficult to measure. The available
anecdotes, however, provide a picture of trafficking that is rampant across Sub-Saharan Africa.
Most women migrants from Ethiopia to the Middle East are transported by either traffickers or
smugglers, and are not recorded in official statistics.9 Martin and others (2002) estimate that
15,000 Malian children between the ages of 12 and 18 had been sold into forced labor on
northern Cote d’Ivoire plantations over a period of a few years, with even greater numbers
pressed into domestic service.10 Black (2004) estimates that 10,000 Nigerian prostitutes work in
over 300 brothels in Europe and South America (although it is not clear how many are
considered victims of trafficking). It became so easy to get Nigerian documents with false
9
Kebede (2001) cites evidence that there are 17,000 women domestic workers in Lebanon, compared with
Ethiopia’s official statistics that 6148 women left Ethiopia form 1992-2001.
10
Parents are paid $30-40 per child, while traffickers provide them with false information concerning working
conditions and pay.
14
information that the Dutch authorities came to view them as invalid until verified. The Institute
for Strategic studies estimated that in 1999 there were as many as 500 organized criminal groups
operating in South Africa, many with African and global networks to smuggle goods and people
(Gastrow 1999).
Conditions that encourage trafficking. It can be hard to distinguish clearly between
trafficking of children and more ‘normal’ migration in West Africa. Migration by children and
child fostering (where children leave the home for a period of years) are frequent and accepted
practices to cope with poverty. In Burkina Faso, 9.5 percent of children from age 6 to 17 do not
live with their parents, and of these 29 percent live abroad, mostly in Cote d’Ivoire (Pizarro
2006). These practices increase children’s vulnerability to traffickers, and can be used to cover
practices that amount to trafficking (Carling 2006). Castle and Diarra (2003) suggest that
focusing on trafficking is problematic in West Africa, where it can be impossible to determine
whether intermediaries have the intention to exploit children.11 Flawed or non-existent birth
registration systems also facilitate trafficking (Wouterse and van den Berg 2004).
Trafficking of women comes in many forms which differ in the degree of coercion
employed. Kebede (2001) describes a range of situations faced by Ethiopian woman who
become victims of traffickers. Often the woman (or young girl) is promised legitimate work but
forced into prostitution on arrival. Or women may not be informed of the terms of contracts that
they are forced to sign (in a foreign language) on arrival, that leave them at the mercy of
employers, or involve large debts to traffickers that have to be worked off before leaving. In the
Middle East, rules that treat women who leave their employers as illegal aliens increase
migrants’ vulnerability to physical abuse.
It is hard to read the anecdotal evidence without concluding that permissive attitudes
towards trafficking and the sexual exploitation of women create an atmosphere that facilitates
such practices. In Nigeria, young girls or their families commit to a written contract and pledge
family assets as security for the debt created when a girl is set up as a prostitute (although it is
not clear to what extent this contract is enforceable) (Carling 2006). In some cases voodoo
rituals are used to bind the family and the girl to the agreement. Prostitution is legal in some of
the sample countries (Burkina Faso, Cote d’Ivoire, Ethiopia, Mauritius and Uganda), which can
make it difficult to enforce laws against trafficking.
Attempts to control. Efforts have been made to prevent child trafficking by governments
and international agencies. Burkina Faso introduced a new travel document in 2002 which
requires the name of the adult accompanying the child and the adult who is to shelter the child,
as a means of fighting trafficking (Wouterse and van den Berg 2004). Save the Children
developed a program in Burkina Faso and Mali to provide children with alternative economic
activities and training in their home villages. In general, addressing the route causes of child
migration is likely to be more successful than attempts to regulate child employment (Delap and
others 2005).
11
Black (2004) notes that NGO and government attempts to return trafficked children to their villages “are met with
unhappiness from children, incredulity from parents, and teasing and humiliation from the children’s peer group.”
15
As a means of protecting women, Ethiopia has strict laws controlling their migration.
Unfortunately, these laws increase vulnerability to trafficking, by forcing women to migrate
illegally and by limiting their ability to call on the authorities for help (Black and others 2004).
Otherwise, the Ethiopian government has not been in a position to offer assistance to its migrants
abroad, and the law against trafficking in Ethiopia has not been implemented properly (no illegal
agent had been convicted of trafficking women outside the country since its enactment—Kebede
2001). And little effort is made to provide pre-departure training or information on conditions of
employment in destination countries that could reduce vulnerability to trafficking. The Nigerian
government has stepped up efforts to strengthen law enforcement, including a systematic
repatriation network for trafficked children, and established a National Agency for the
Prohibition of Trafficking in Persons in 2003 (Carling 2006). However, according to Black
(2004) the U.S. Department of State still argues that Nigeria does not fully comply with
minimum standards for the elimination of trafficking. In South Africa, the absence of domestic
anti-trafficking legislation offers law enforcement little incentive to pursue criminal syndicates
(Martens and other 2003).
Impact of immigration on sample countries
Very little hard data exists on the economic impact of immigration on the sample
countries. Economic analyses of the implications of migration for the low-income African
countries appear to be unavailable. Some information, albeit extremely limited, anecdotal, and
in some respects contradictory, is available on immigrants in the labor market for two middleincome countries in the sample. In Mauritius, immigration is low (1.8 percent of the population
in 2003), and broadly accepted. It appears that immigration has helped boost employment and
the economy: foreign construction workers have been essential to the timely completion of
infrastructure projects, and it is generally accepted that several EPZ factories would have to shut
down without guest workers (Hein 2004).
In South Africa, by contrast, the impact of immigration on labor markets has been the
subject of great controversy. Studies of sectors with large numbers of irregular migrants show a
consistent violation of labor standards and wages below the legal minimum (Black 2004), while
Human Rights Watch (2006) claims that workers on commercial farms, documented or not, are
subject to violations of basic employment law protections. Unions also report that
undocumented migrants working on farms may be paid shelter and a plate of food a day, with
their main concern the undercutting of wages for natives (Solomon 1996). On the other hand,
mining interests (including post-apartheid) have benefited greatly from the immigration of
workers, with annual average inflows totaling 300,000 in the early 1970s and 200,000 in the
1980s and early 1990s (UN 2004). Cross-border traders and other small businesses owned by
foreigners play an important role in South Africa’s informal economy, and have engendered
much resentment from local traders. Perbedy and Rogerson (2000) report incidents of violence
and intimidation practiced against non-South African street traders by gangs claiming to
represent the unemployed.
16
Remittances
Remittance transfers from emigrants to their origin countries have attracted increasing
attention over the past few years, in part due to the sharp rise in official data on remittances
receipts by developing countries.
Size of remittances
The size of remittance inflows varies considerably among the sample countries.
According to official statistics, Nigeria receives the most in dollar terms (almost $2.4 billion in
2006), while Uganda, South Africa and Senegal receive more than $500 million (figure 5). For
most of the countries, official remittances were less than 5 percent of GDP in 2006, with the
exception of Lesotho (25 percent), Uganda (8 percent), and Senegal (7 percent). Remittances
have increased significantly over the past few years. For the sample countries as a group,
remittances inflows rose from 0.4 percent of GDP in 1980 to 1.3 percent in 2005. Excluding
Uganda (which lacks data before 1999), remittance receipts in dollar terms more than tripled
between 1994 and 2006. It is unclear how much of the increase in recent years represents better
reporting or greater willingness to use formal channels after stricter anti-money laundering rules
in the wake of the terrorist attacks of September 11, 2001. Also, the dollar value of remittances
to the Francophone countries has increased since 2000 due to the appreciation of the euro.
Table 2: Remittances to sample countries, 1980-2005
(percent of GDP)
Burkina Faso
Ethiopia
Ghana
Lesotho
Mali
Mauritius
Nigeria
Senegal
South Africa
Uganda
total
1980
7.8
1985
7.8
0.1
0.1
77.3
5.1
..
0.0
61.0
3.3
..
..
0.0
2.6
0.1
..
1995
3.2
0.4
0.3
44.1
4.5
3.5
2.9
3.3
0.1
..
0.0
3.1
0.1
..
0.4
1990
4.5
0.0
0.1
69.5
4.4
0.0
2.5
0.1
..
0.5
..
0.6
0.9
2000
2.6
0.7
0.6
29.3
3.0
4.0
3.3
5.3
0.3
4.0
1.4
2005
1.0
1.2
0.9
24.4
3.0
3.3
2.3
6.1
0.3
7.4
1.3
Source: DEC Prospects Group, the World Bank
17
Figure 5: Remittance receipts by the sample countries
Estimated Remittance Inflows, 2005
Estimated Remittance Inflows, 2005
(US$ millions)
(percent of GDP)
Uganda
Uganda
South Africa
South Africa
Senegal
Senegal
Nigeria
Nigeria
Mauritius
Mauritius
Mali
Mali
Lesotho
Lesotho
Ghana
Ghana
Ethiopia
Ethiopia
0
500
1000
1500
2000
2500
0
5
10
15
20
25
30
Sources: DEC Prospects Group, the World Bank
It is likely, however, that the official statistics considerably understate the true size of
remittance receipts. Estimates of unrecorded remittances vary widely by country, but many
writers commenting on the subject believe that official figures capture less than 50 percent of the
total amount transferred (see below). By its nature the informal remittance market is difficult to
observe. But the anecdotal evidence in favor of significant understatement of the size of
remittances is strong.
Economic impact
Undoubtedly some areas of the sample countries are heavily dependent on remittances,
and have been for a long time. A survey of households in the Senegal River valley (which
crosses Mali, Senegal and Mauritania) showed that migrant remittances provided 65 percent of
cash income, probably including remittances from internal migrants (Findley and Sow 1998).
Poverty impact. Several studies have attempted to measure the impact of remittances on
poverty or expenditures. Care has to be taken in reading these studies. Some simply measure
the increment to household income from receiving remittances. Remittances are thus manna
from heaven, and no attempt is made to take into account what the migrant would have earned
had he or she remained at home. For example, Lichtfield and Waddington (2004) find that,
controlling for other determinants of income, Ghanaian households whose head had migrated
spent on average $78 per adult equivalent per annum more than households whose head had not
migrated. By 1998/99, households with migrated heads spent only $39 more than non-migrants.
Gustafsson and Makkonen (1993), based on a 1986/87 household survey, find that remittances
are the main source of income for 35 percent of Lesotho households, and that in the absence of
remittances consumption per capita would fall by 35 percent.12 This perspective is appropriate
for explaining the current welfare of households, but not for determining whether households are
better off for deciding to send a migrant abroad.
12
They make the extreme assumption that migrants return home but are unemployed and make no contribution to
household income, thus further reducing per capita consumption.
18
A few studies do incorporate estimates of what migrants would have earned had they
stayed at home, and thus provide a more accurate perspective on the impact of migration
decisions. Lauchaud (1999) estimates that international remittances to Burkina Faso reduce the
headcount measure of poverty by 7.2 percentage points in rural areas and 3.2 percentage points
in cities. The fall in the headcount ratio is statistically significant only for subsistence farmers,
for the unemployed, and for the self-employed in urban areas. Adams (2006) finds that
remittances reduce the level, depth and severity of poverty in Ghana. Poverty is reduced by more
when international, as opposed to internal, remittances are considered, and when more sensitive
poverty measures (the poverty gap and squared poverty gap) are used. International remittances
do result in a small rise in the headcount level of poverty, indicating that for income deciles near
the poverty line, remittances are slightly smaller than the estimated earnings of migrants had they
not left. However, international remittances have a larger impact in reducing the poverty gap,
indicating that a significant share of remittances go to households that are well below the poverty
line (which doesn’t necessarily affect the headcount ratio but does improve the welfare of the
poor). Gubert (2005) notes that studies of the impact of remittances typically do not account for
multiplier effects. She gives the example that market gardening would probably not have
become so popular in Mali if smallholders had not been sure of a demand for their produce.
Consumption smoothing. Remittances may also help to smooth consumption in the face
of economic shocks. Quartey and Blankson (2004) find only weak evidence of consumption
smoothing for all households from the Ghana Living Standards Measurement Survey, but a very
pronounced role in consumption smoothing for food crop farmers, who are the poorest group.13
Addison (2004) finds that remittances to Ghana were less volatile over 1990-2003 (standard
deviation of .21) than ODA (.60) or FDI (.61). A survey of the Senegal River valley found that
remittances were named as an important source of money for emergency shortfalls 14 percent of
the time in an average year, but 27 percent of the time during drought (Findley and Sow 1998).
In a study of Ethiopia, Areda (2005) finds that the most vulnerable households (using as a proxy
female-headed households) and those most subject to income variability rely on remittances for a
significant share of income.
Expenditure allocation. Analyses of the allocation of expenditures financed by
remittances often suffer from two weaknesses. First, the data are suspect, as individual reports of
how remittances are spent may not reflect the likely difference in expenditure allocation with and
without remittance income (some authors rely on reported expenditures, while others compare
expenditure patterns of similar households that receive remittances with those that don’t).
Second, it is not clear what conclusion to draw, even if reliable data on how expenditures are
affected by remittances are available. Presumably households are the best judges of how they
should spend their extra income.14 Asiedu (2004) cites a Ghanaian study where 70 percent of
remittances were for recurrent expenditure, and less than 30 percent invested. Palmer (1985)
finds that women, (especially younger women) that received remittances favored the education
of daughters over sons, on the grounds that daughters were more loyal to natal families than
13
No attempt is made to estimate the income that migrants would have earned if they had stayed at home. Doing so
could have changed the results, as larger households might have a greater potential for risk diversification.
14
One way of looking at this issue is that analysts generally don’t spend a lot of time determining whether
domestically earned wage income is worthwhile on the basis of how it is spent.
19
sons.15 Birdsall and others (1988) suggests that the availability of remittances increases the
demand for health services and has a significantly positive effect on the probability of seeking
modern care.16 Sander and Barro (2003) report surveys of Senegalese households showing that
remittances are mostly used for current consumption, with only 10 percent of remittance receipts
devoted to savings, and a small portion allocated to housing expenditures.
Impact on growth. Several writers observe that areas that receive large amounts of
remittances don’t appear to have achieved a process of self-sustaining growth. Martin and others
(2002) claim that the large volume of remittances to the Kayes region in Mali has improved the
lives of residents and financed the construction of public institutions, but “do not seem to have
led to the establishment of large numbers of businesses that promise stay at home development.”
Similarly, Azam and Gubert (2005) found that the receipt of remittances reduced work effort in
Kayes. While households receiving remittances did adopt new technologies, agricultural
performance was worse than in nonmigrant households. On the other hand, remittance income
may be large enough to adopt technologies that result in increased production. Findley (1989)
reports that remittances made irrigation possible in the Senegal River valley.
For Lesotho, Lucas (1987) shows that emigration to South Africa’s mines diminishes
domestic crop production over the long term (unlike in four other Southern African countries,
where he finds that emigration to the mines enhances crop productivity and cattle accumulation
in the long run). He ascribes the difference to massive soil erosion in Lesotho from the neglect
of crop production owing to emigration. Russell and others (1990) quote a World Bank Staff
Appraisal Repot attributing the stagnation of agricultural production in Lesotho and the sharp
decline in area under cultivation to the receipt of remittances. By contrast, Mochebelele and
Winter-Nelson (2000) find that in Lesotho, farms with migrants abroad that produced cereals
were more efficient than non-migrant farms, although there was no significant difference
between migrant and non-migrant households producing other crops.
The examples of the Kayes region of Mali and Lesotho point to a threshold effect. It
may be that a significant, negative impact on work incentives can be found only in regions and
countries with truly massive levels of remittances. In 1990, remittances from South Africa
accounted for 70 percent of GDP in Lesotho, while Plath and others (1987) note that about 60
percent of the male labor force were absent, preventing up to half of the available acreage from
being cultivated.
One issue is the extent to which emigrants can, and desire, to condition their provision of
remittances on work effort or the allocation of remittance expenditures. To the extent that such
conditions are effective, it is possible that remittances would be more productive. There is some
evidence that migrants’ decision-making power in the management of household affairs is being
boosted by advances in telecommunications. For example, emigrants from Senegal are using the
services of companies that specialize in the long-distance sale of foodstuffs to cover family
expenses, pay phone connections, and ensure payment of bills, thus exercising greater control
over what expenditures are financed by remittances (Tall 2003).
15
16
Quoted in Russell and others (1990).
Quoted in Russell and others (1990).
20
Increasing access to finance. Some developing countries have increased their access to
international financial markets by securitizing future flows of remittances. Essentially these
transactions allow countries to borrow by assigning to the creditor a portion of future foreign
exchange received through remittances. Countries which receive a large amount of remittances
through official channels can use such transactions to gain access to markets that would
otherwise be closed to them, and to improve the terms on loans. For example, a Ghanaian
government bank that had been involved in transferring remittances (in cooperation with a
money transfer agency) borrowed $40 million in the late 1990s by securitizing future remittance
receipts. This arrangement allowed the bank to raise a larger sum of hard currency financing, for
a longer maturity and at a relatively low interest rate, compared to what it could have done
through unsecuritized borrowing (UNCTAD 2006).
Use of remittances to finance public goods
Organizations of migrants dedicated to mobilizing funds from the diaspora and financing
public projects in origin countries are common in Europe. For example, Gubert (2005) reports
about 1000 such organizations in France, one third of them dedicated to Mali, Mauritania, or
Senegal. Migrants in France from the same area of Senegal set up “secondary villages” to
participate in, or assume responsibility for, the building and furnishing of schools (Russell and
others 1990).
Areas that are important sources of migrants have received significant benefits from
diaspora organizations, particularly in the form of public works. For example, in Yelimane, a
city in Mali, a disapora organization accounted for 180 wells and boreholes, 70 schools, 11
dispensaries and 19 co-operatives, for a total expenditure of CFAF 7 billion (Gubert 2005).
AFFORD (2000) quotes estimates that over 60 percent of the infrastructure of Kayes can be
attributed to the diaspora in France. There are numerous Nigerian hometown and alumni
associations, which represent professional interests as well as channeling remittances (Black
2004)
One issue with the efficiency of expenditures by village associations is that migrants are
often prominent in developing ideas for projects, but may not have necessary skills and expertise,
and often feel marginalized owing to distance. At the same time, local officials may feel
disempowered by projects, which may underline their lack of capacity to provide for their
community (van Vlaenderen and others 2004)
Diaspora associations may provide technical advice instead of, or in addition to, money.
Sall (2005) describes organizations such as the Association of Engineers for the Development of
the Sahel, which provides support and advice for some 250 project in the Kayes region of Mali.
Transactions costs and modalities
There is a considerable variety of remittance services available to migrants. Formal
channels typically include banks, post offices, and money transfer operators such as Western
Union and Moneygram. Informal channels may involve bringing money home by hand; bringing
goods rather than cash; sending cash with friends, relatives, or other carriers (in East Africa,
some courier and overland bus companies provide money transfer services—Sander and others
2001); and depositing money with unregulated individuals in destination countries that arrange
21
for distribution in origin countries. This last modality takes advantage of modern
communications facilities to bundle remittances, and thus economize on transfer costs. For
example, Sengalese migrants in northern Italy deposit money with a transfer agent, who gives the
migrant a code number. The agent communicates information on the transaction by telephone to
his representative in Senegal, and the migrant does the same to his or her family (Fall 2003).
Modern telecommunications technology also is playing an increasingly important role in formal
transfer networks (Tall 2003). In South Africa cell phones are now being used to transfer money
(Gupta and others 2006), and the post office is offering a PIN money order which provides
immediate access to funds on line to any post office in the country, while use of the internet,
ATM machines and prepaid cards is increasing (Genesis 2003).
There also are legal means for migrants to spend money for the benefit of households in
origin countries, rather than sending cash directly to households. For example, some migrants
purchase state bonds, where income is paid out to the household in the origin country, or
purchase health insurance that covers relatives at home (Higazi 2005). Senegal Conseils (in
Lyon) allows migrants in France to purchase goods and have them delivered to the family in
Senegal (Sander and Maimbo 2003)
Remittances in the form of goods, or directed purchases of domestic goods, can also be
important. Sabates-Wheeler and others (2005) cite a Ghanaian study showing that the majority
of remittances were brought back by returnees in the form of goods. In Ethiopian survey data
covering four samples from 1994 to 2000 (which includes internal remittances), the share of
goods in total remittances averaged 41 percent, with this ratio declining over the period owing to
the increasing availability of imported goods in domestic markets (Aredo 2005). This practice
probably is encouraged by allowing migrants to bring in personal effects, including a range of
household items and a car for private use, free of tax and tariff (Asmelash 2006). A 1995 study in
Ghana (that is difficult to believe) claims that 95 percent of remittances were in goods, including
vehicles, appliances, equipment and machinery, rather than cash (Anarfi and others 2003).
According to Higazi (2005), this estimate was made before the Bank of Ghana implemented
improvements in the monitoring of remittances, so that the share of cash transfers could have
increased. The distinction made in the data between goods brought home by migrants and
standard imports is also unclear. The Ghanaian housing industry is so dependent on remittances
that most housing estates quote prices in dollars and have accounts overseas into which
emigrants can make payments to purchase housing.
The distribution of remittances between formal and informal channels varies
considerably across countries, although empirical studies disagree on the relative importance of
different channels. A study of Sahelian migrants in France found that one-third sent remittances
by hand, one-third by post office and only 6 percent through banks (Russell and others 1990).
Gubert (2005) reports a survey of Mali and Senegalese emigrants where 56 percent sent money
through an intermediary visiting the home country, and 15 percent used postal orders. Bank
transfers tended to be used for larger amounts, and postal orders for smaller amounts. Anarfi and
others (2003) claim that the majority of transfers to Ghana go through formal routes (banks, post
office, or Western Union), and that the frequency of remitting is positively related to the
likelihood of using a formal channel. Nevertheless, a substantial number of migrants did send
cash back or brought it on visits. By contrast, preliminary data from the Ghana Transnational
22
Networks research program indicate that unregistered remittances may account for as much as 65
percent of total remittances (including internal remittances) in Ghana. Tiemoko (2004) finds that
migrants who remit less frequently, and less-skilled migrants, are more likely to use informal
channels. Tall (2003) reports that over half of the remittances to Senegal come through channels
other than banks or the post office. CGAP (2003) reports that focus group discussions, data from
Finscope South Africa, and BNLS surveys show that taxi drivers or friends and relatives capture
53 percent of cross-border remittances out of South Africa, the post office 32 percent, banks 5
percent and others 10 percent.
Some writers cite policies that impede migrants’ use of formal channels. In South Africa
money transfer agencies are required to have a banking license, to invest in an expensive
exchange control reporting system, to ensure payment of taxes on transmitted funds, to ensure
that the funds are not proceeds from crime, and to verify that the sender is a resident, has an
immigration or work document authorizing her to earn rands, and is not in breach of exchange
control laws. This array of requirements limits competition and raises fees. It also encourages
senders to turn to informal networks, and money transfer agents to refuse to serve low-income
individuals whose bona fides may be difficult to verify (Gupta and others 2006; Genesis 2003).
There has been some discussion as to whether microfinance agencies could become more
involved in remittances. Their relatively poor clientele and low cost structure could help reduce
the cost of remittance transfers and encourage poorer migrants to send remittances through
formal channels. However, Higazi (2005) does not find much evidence of microfinance
involvement in remittances transfer agents in Ghana. By contrast, Sander and Barro (2003) find
that microfinance institutions are cooperating with banks to offer transfer services in Senegal.
The regulatory framework that permits this can be cumbersome, as authorization from the
Ministry of Economy and Finance is required, and the microfinance institution must have
sufficient capital and financial expertise. The Uganda Microfinance Union introduced a pilot
money transfer product in 2001, although its clients tend to be larger corporations, due to high
fees (Black and others 2004).
Scattered anecdotal evidence indicates that fees are lower for informal networks than
banks and money transfer operators (although not necessarily for postal orders), at least for small
amounts (Higazi 2005). Loup (2005) claims that for Malian immigrants in France, informal
transfer services are one-third the cost of Western Union. Gupta and others (2006) report that
the cost of transferring 250 rand was lowest when a friend or taxi driver simply took the money
across the border, and highest when banks were used. While post office transfers in South Africa
were competitively priced, they were neither fast nor secure. Fall (2003) for Senegal, CGAP
(2003) for South Africa, and Gubert (2005) and Sander and Maimbo (2003) for Mali and Senegal
also cite the slowness, and in some cases lack of reliability, of transferring money through postal
orders. Fees for remittances in South Africa differ dramatically by modality: a bank draft may
cost 35-68 percent of the principal, electronic bank transfers 19-62 percent, postal orders 8
percent, Moneygram 25 percent, and an online transfer service 6 percent (Gupta and others
2006).
Fees are typically smaller, as a percentage of the amount transferred, the larger the
amount. For example, to send money to Ghana, money transfer operators in the United Kingdom
23
charge a maximum of 12 percent for 100 pounds and of 7 percent for 500 pounds (Gupta and
others 2006). Sander and Maimbo (2003) report that the fee charged by Western Union and
Moneygram to send money from Uganda declines from between 6 and 35 percent for 100,000
shillings (about $60) to between 2-6 percent for 1,000,000 shillings (approximately $600).
Skilled migration
The emigration of skilled professionals from African countries has been cited by
numerous authors as an important constraint on development. Nevertheless, it is extremely
difficult to measure the impact of skilled emigration on origin countries. In theory, skilled
emigration may impair development because: (i) skilled professionals generate benefits, for
example training of colleagues or strengthening governance, that are not fully reflected in their
salaries; (ii) firms may benefit from economies of scale, and skilled professionals (e.g. computer
scientists) may be required to achieve large size; (iii) if the education of skilled emigrants is
funded by the state, then emigration imposes a fiscal cost; and (iv) emigration will increase the
price of services that require technical skills (e.g. health care), where it is often difficult to
substitute low-skilled workers (World Bank 2006). Unfortunately, it is extremely difficult to
measure the first two effects, and lack of data constrains estimates of the last two for African
countries.17 A great deal of anecdotal evidence is available on skilled emigration by health
professionals, however. We will first consider estimates of the overall rate of skilled emigration,
and then turn to whatever sectoral information is available.
Data on skilled emigration
The share of skilled emigrants in the total number of trained professionals varies among
the 10 countries, but is quite high for some. Aggregate data are available on college-educated
people from the sample countries that reside in OECD countries, as a share of total collegeeducated people in the origin country (figure 6). These data are collected from destination
country sources, given the spotty availability and lack of quality of data on skilled emigration
from African source countries.18 Even so, the data suffer from several limitations, including the
poor quality of reports in the origin countries and the fact that it is impossible to tell how many
of these highly-educated people were educated in the OECD country. Nevertheless, they can
provide some view of the importance of overall highly-skilled emigration.
According to these data, the flight of highly-educated persons has been most severe in
Mauritius (57 percent of college educated natives residing in OECD countries), Ghana (48
percent), and Uganda (37 percent). Senegal (18 percent) and Mali (15 percent) occupy an
intermediate position, while in South Africa, Lesotho, Burkina Faso, Nigeria, and Ethiopia, the
17
Konseiga (2005c) does present model-based estimates of the growth impact of skilled emigration from Burkina
Faso, which finds that emigration to industrial countries improves growth prospects, but emigration to Cote d’Ivoire
doesn’t. OECD (2003) estimates that the educational costs of South African health professionals working in other
countries may equal $1 billion, although this does not account for remittances or the possibility of return. Van
Rooyen (2003) estimates that the brain drain costs South Africa 2.5 billion rand a year.
18
For example, Bhorat and others (2002) claims that the South African Department of Home Affairs
acknowledges the underestimation of the extent of skilled emigration, and Buchan and others (2003) believe
it is impossible to determine the actual number of nurses leaving South Africa.
24
rate is at or below 10 percent. Unfortunately, it is impossible to say what level of highly-skilled
emigration is critical for growth, nor how much the sectoral allocation matters (and what it is).
Figure 6: The brain drain
Highly-Educated Emigration to OECD Countries, 2000
(percent of highly-educated in source country)
Uganda
South Africa
Senegal
Nigeria
Mauritius
Mali
Lesotho
Ghana
Ethiopia
Burkina Faso
0.00%
10.00% 20.00% 30.00% 40.00% 50.00% 60.00%
Source: Docquier and Marfouk (2004)
Much has been written about the outflow of doctors and nurses, and some of the statistics
are quite alarming. For example, 69 percent of doctors trained in Ghana from 1995-2002 left the
country, while estimates of vacancy levels in the Ghanaian Ministry of Health exceed 65 percent
for doctors and nurses (Bump 2006). However, it is unlikely that the Ministry has the resources
to hire up to their full staffing levels. Bhorat and others (2002) describe how the attractiveness of
higher salaries and better working conditions in South Africa has increased vacant nurse posts in
Lesotho. OECD cites the South African public sector’s large shortfall in nurses, in part due to
rising emigration. A recent dataset, based on census data in receiving countries, shows that 21
percent of physicians, and 12 percent of nurses, born in the 10 countries are working in the 8
industrial countries for which data are available. This includes African professionals trained in
the industrial countries. The countries most affected by the health professionals’ brain drain are
Ghana, Mauritius, and Senegal.
25
Table 3: Health professionals working in industrial countries
(professionals from each country working abroad as share of total professionals, working
domestically and abroad; percent)
Burkina Faso
Ethiopia
Ghana
Lesotho
Mali
Mauritius
Nigeria
Senegal
South Africa
Uganda
Total for 10 countries
Physicians Nurses
19.9
2.4
29.3
16.8
54.6
24.1
6.6
0.9
22.9
15.0
45.5
63.3
13.1
11.7
51.3
26.9
19.2
4.2
40.6
10.1
21.0
11.7
Receiving countries include Australia, Belgium, Canada, France, Portugal, Spain, United Kingdom and United
States.
Source: Clemens and Petterson (2006)
A few writers have provided information on skilled emigration in other professions.
However, these tend to be either numbers of skilled emigrants without information on the
demand for these workers in the origin country, or unquantified references to large skilled
emigration or vacancies. It is estimated that Ghana lost between one half and two-thirds of its
most experienced, top-level manpower between 1960 and the mid-1980s (Rado 1986). EU
(2006) provides survey evidence that 43 percent of firms in Mauritius believe that the skill level
of the workforce is a serious obstacle to their operations; the level of skills is given as the second
most important constraint for service sector firms. Black (2004) claims that the exodus of
Nigerian academics and students reflects, and reinforces, a decline in educational standards in
Nigerian universities. When Nigeria’s oil-driven prosperity ended with the fall in oil prices in
the early 1980s, the country saw an outflow of highly-skilled personnel to the oil-rich countries
of the Middle East. In turn, Fadayomi (1994) notes the considerable migration of skilled
personnel to Nigeria from other West African countries. Bhorat and others (2002) note the high
vacancy rates at the top end of the labor market in South Africa, and claims that South Africa
lost 13 percent of skilled hires per year over the 1990s. By contrast, McDonald and Crush
(2002) cite survey data indicating that employers of skilled personnel in South Africa do not
appear to have major problems in hiring skilled workers, with the notable exception of the
information technology sector. Uganda is reported to have lost more than half of its professional
and technical workers during the rule of Idi Amin from 1971-79 (Russell and others 1990).
It is reasonable to conclude that skilled emigration does have a negative impact on the
provision of some skilled services, and in particular public health care, in most of these
countries. Buchan and others (2003) claim that the emigration of nurses from South Africa
demotivates remaining nurses by increasing their workload and degrading the overall quality of
service. However, given the many problems (low salaries, overburdened staff, and poor
facilities) facing these health care systems, the contribution of emigration can be difficult to
measure. For example, despite high quoted vacancy rates for nurses in the Ghanaian Ministry of
26
Health, De Souza (2006) claims that a lack of infrastructure and inadequate investment in the
health sector mean that Ghanaian nurses who have qualified abroad are unable to find
employment back home. Health professionals may be leaving South Africa in droves, but
35,000 registered nurses are either inactive or unemployed, given the limited funds available for
public health (OECD 2003). Moreover, weaknesses in health care systems tend to increase
skilled emigration. Buchan and others (2003) cite focus group interviews showing that long
hours, poor resources, and a high ratio of patients per nurse are major reasons for emigration
from South Africa.
Determinants of skilled emigration
Given the huge salary differentials between the pay of skilled professionals in Europe
and America versus in Africa, the desire for many skilled professionals to emigrate is not exactly
a puzzle. For example, a skilled nurse earns about $11,000 in public service in South Africa,
$22,000 in the United Kingdom, as much as $35,000 in Saudi Arabia, and over $40,000 in the
United States (OECD 2003).19 Nevertheless other drivers of skilled migration cited in the
literature include social pressures owing to large families and a hierarchical structure that may
constrain the ambitions of the young professionals (Thombiano 2000 on Burkina Faso), the
availability of advanced facilities, and professional contacts developed when studying abroad.
Skilled emigration from South Africa may be boosted by issues that go beyond the workplace.
Van Rooyen (2001) and Mattes and Richmond (2000) provide survey evidence that violence is
boosting emigration (50 percent of South Africans who formally indicate an intention to
emigrate report in exit polls that are leaving because of violent crime—Myburgh 2004). On the
other hand, the idea that racism and affirmative action are a major cause of increased emigration
rates from South Africa is not supported by survey evidence (McDonald and Crush 2002).
Skilled emigration may reflect other issues in the educational system, or poor economic
policy. Russell and others (1990) state that concerns over the emigration of skilled workers from
Ghana reflect an excessive expansion of education. Fall (2000) claims the scarcity of trained
professionals in Senegal is due to the low share of students in technical fields, with literature and
social science more common. This may reflect the poor investment climate in Senegal, as a large
share of science graduates choose to work in teaching and government administration, rather
than the private sector. Similarly, Bhorat and others (2003) cite a growing number of graduates
from South African universities in areas where the demand for skills hasn’t grown (e.g. home
economics, philosophy, religion, language) while business degrees are stagnating and
engineering graduates have declined.
Overall, the impact of skilled emigration on the sample countries is impossible to
determine. While the data on the health sector does raise serious concern, in other sectors little
can be found on outflows of trained professionals relative to the demand for their services. It is
useful to note that the Mauritius, the country with the highest ratio of college-educated people
residing in OECD countries, was the only country (of these 10) in a recent UN survey to state
that they were interested in boosting emigration rates.
19
Note, however, that these data are not adjusted for differences in price levels. A nurse earning $11,000 in South
Africa can purchase goods valued at almost $30,000 in international dollars (adjusted for differences in purchasing
power parity), or about 70 percent of the purchasing power of a public sector nurse in the United States (by the
figures given in OECD 2003 and in the World Bank’s World Development Indicators.)
27
Whatever the impact of skilled emigration, few writers have suggested policies to
prevent it. In the late 1980s, Uganda imposed foreign exchange restrictions and clearance
requirements for foreign travel on professionals and civil servants (Russell and others 1990), but
these were largely dismantled by 2000. Outright prohibitions of skilled emigration are rarely
proposed. In part this is due to the strong presumption in international discussions of human
rights, as well as in the legal framework of the sample countries, of the right to emigrate. Such
limits would also be difficult to enforce, and potentially counterproductive as they would
discourage return. Some attempts are made to ensure some return on government investment in
training professionals. For example, South Africa requires graduating doctors to complete one
year of service in rural areas of the country prior to registration (a requirement to practice). So
far only 8 percent of graduates have dropped out, either by deferring their year of service or
refusing to register (OECD 2003). Another approach to overcoming the shortage of skilled
professionals is to attract skilled immigrants, although of our sample country this could make a
substantial contribution only in the middle-income countries (South Africa and Mauritius).
Despite South Africa’s relatively hostile immigration policy, Bach (2003) reports that in 1999
almost 80 percent of rural doctors were immigrants. Indeed, South Africa is changing its
immigration laws to attract more skilled workers by making it easier for skilled persons to obtain
extended work permits or permanent residency (McDonald and Crush 2002). However, any
success that South Africa may have is likely at the expense of other, particularly nearby,
countries in Sub-Saharan Africa.
Efforts to encourage return and circular migration
Developed destination countries have provided subsidies to encourage return, with mixed
results. For example, in the 1980s France financed an assisted return program for agricultural
cooperatives in the Kayes region, which were still functioning as of 2002 (Martin and others
2002). A program to provide loans to migrants in the 1980s to help them set up businesses in
Senegal, however, had only 10 recipients, and five were out of operation six years later, with the
other five virtually bankrupt (Diatta and Mbow 1999). A later French program paid about 500
unauthorized Malians in France to return voluntarily in exchange for CFA 2.5 million ($3,600)
and open businesses, 80 percent of which were still operating after 2 years. However, Sall (2005)
cites administrative delays and difficulties in establishing businesses in Mali as hampering these
returnees.
Origin countries also are involved in efforts to encourage the return of skilled emigrants
and to foster ties with the diaspora. In Ethiopia, special travel documents allowing visa-free entry
are issued to foreign citizens of Ethiopian origin and their spouses (IOM 2005a). The Nigerian
government made significant efforts to contact its émigré professionals, while the Senate
abolished a measure that would have meant that Nigerians who became citizens of other
countries would lose their Nigerian citizenship (Black 2004). Mali, Senegal, and Burkina Faso
established government ministries or agencies that encourage cultural exchanges, the provision
of information, and investment in local businesses by the diaspora (Ammarrasi 2005). The
Senegalese government encouraged large inflows of savings from Gabon and Cote d’Ivoire by
sending missions to inform Senegalese émigrés of opportunities in the home country (Diatta and
Mbow 1999). Ghana increased access to dual citizenship for overseas Ghanaians in 2002
(Anarfi and others 2003), and has attempted to increase the interest of the diaspora in Ghana
28
through invitations to meetings in Accra (Bump 2006). South Africa invited skilled
professionals in the diaspora to sign up at a website to help train South African workers or
students, or assist South African firms with research, business contacts, and technology transfer
(Mutume 2002). Uganda has sent government speakers to meetings of the diaspora to “raise
awareness of the development objectives of Uganda” (Black and others 2004).
Black and others (2003b) find that government incentives were negligible in influencing
return, as very few Ghanaian returnees benefited from, or even knew about, incentives for return
(they cite a few instances of incentives for return from the Ghanaian Tourist Board and the
Ghana Private Road Transport Union). Mauritius introduced the Scheme for Attracting
Professionals in Emerging Sectors in 2002 to attract information and communications
technology professionals to return. However, cumbersome administrative procedures resulted in
only 10 approvals of 29 applications, of which 5 actually materialized (EU 2006).
International organizations have also encouraged return. The Migration for Development
in Africa (MIDA) program (managed by the International Organization for Migration) attempts
to match the know-how requirements in Africa with the skills of volunteer migrants from African
countries. The United Nations Development Program provides funds to encourage at least
temporary return (referred to as the Transfer of Knowledge through Expatriate Networks). For
example, TOKTEN paid 133 Malians to return as consultants to teach and do research.
Migrants may become more productive from their overseas work. A survey of 304
returning migrants to Ghana and Cote d’Ivoire in 2000-2001 found that most had increased their
education or professional experience, and a large majority felt that the knowledge obtained
abroad was important, or very important for their current work (Ammarrasi 2005). Black and
others (2003a) cite survey data showing that 90 percent of highly-skilled returnees and 40
percent of less-skilled returnees hoped to apply what they learned abroad.
Health issues
Migration often serves as a vector for disease, in part because migrants are often
vulnerable to disease and may have little access to health services, and in part because migrants
are carriers of disease between communities.
The role of migration in spreading AIDS throughout Africa has been well documented.
Migrant communities tend to have a higher than average rate of HIV infection, as mobility can
encourage or make people vulnerable to high-risk sexual behavior (and in many countries the use
of protection against sexually-transmitted disease is limited), while making them more difficult
to reach for testing and preventative measures (Black 2004). For Sub-Saharan Africa as a whole,
IOM (2005b) shows a strong relationship between HIV prevalence among pregnant women and
the proportion of new immigrants. Several studies have tied migration to increased HIV/AIDS
in the sample countries:
(i) Lesotho has an overall HIV-positive rate believed to be highest in world, and is the only
place in Southern Africa in which more men than women are HIV positive, the result of miners
living in single-sex hostels w/access to sex workers (Cobbe 2004);
29
(ii) in Mali, the rise in HIV-positive cases and the high growth rate of new AIDS cases was
partly due to migration from Cote d’Ivoire and Senegal, where sero-prevalence rates were
increasing at a rapid rate (World Bank 1993);
(iii) in some parts of Senegal, AIDS is referred to as Cote d’Ivoire fever, as the first AIDS
victims were returning migrants from that country (Fall 2003);
(iv) in a study of Matam, a city in northern Senegal, migrants to other African countries face a
greater risk of HIV infection than less mobile populations, and focus groups and personal
interviews confirmed the import of HIV/AIDS by migrants (Thiam and others 2004);
(v) a study in Kwazulu/Natal in South Africa found that HIV prevalence among migrants and
their partners was 24 percent, compared with 15 percent among non-migrants and their partners
(Lurie 2004); and
(vi) a longitudinal cohort study in the Masaka District of Uganda found that age and sexstandardized HIV-positive rates were 5.5 percent for adults who had not moved from home,
versus 11.5 percent for those who had left the area (Nunn and others 1995).
AIDS is not the only disease that has been affected by migration. In Burkina Faso, an
immunization program against measles achieved a high coverage rate, but was followed by a
large outbreak of measles among the target population. This was the first reported failure of a
widely successful strategy for controlling measles. On investigation, it became clear that the
immigration of children from Cote d’Ivoire, who had not been immunized, played a major role in
the failure of the immunizations to interrupt the transmission of measles (Yameogo and others
2005). Forced migration to different ecological zones can expose populations to diseases for
which they lack immunities. For example, refugees from the highlands areas of Ethiopia
developed a high prevalence of malaria after settling in camps in malarial-endemic areas of the
eastern Sudan (Russell and others 1990). They also cite evidence that workers returning from
the mines in South Africa suffered disproportionately from tuberculosis and other communicable
diseases, that transmission of the Guinea worm in Nigeria was linked to migration, and that
refugee camp conditions in Uganda resulted in partially-treated tuberculosis cases becoming
rapidly infectious.
Priorities for research for the Africa project
The economic literature on migration in Africa suffers from a dearth of information.
Official data on the numbers of people migrating and the amount of remittances are likely
seriously understated, while analyses of the economic impact of migration on both destination
and source countries are largely speculative. In the countries covered by this paper (with the
exception of Mauritius), virtually all aspects of migration are poorly understood, and government
efforts to influence migration to improve welfare have often been impotent or counterproductive.
Thus there is no shortage of areas where the generation of new information through the Africa
project can improve understanding. The limited resources and techniques available for gathering
information make it important to set priorities for future work. Here I will highlight a few areas
where this paper can provide guidance.
30
Some progress in quantifying the number of migrants, as well as their economic and
geographical distribution, could help policymaking. At a minimum, better information might
elevate the debate in South Africa over the impact of irregular migration, or lead governments in
the poorer countries to consider migration policy before a crisis, perhaps helping to avoid the
immense human suffering occasioned by past expulsions of foreigners.
There is little evidence that efforts by African governments to harness the diaspora have
had much impact. Survey information would be useful in gauging the awareness of, and
participation in, these programs by emigrants. There have been considerable private efforts to
distribute funds for local infrastructure projects in areas that are the major sources of emigrants.
Some exploration of how the government might cooperate with these private initiatives to
channel information, technical assistance, and business contacts to local entrepreneurs might be
useful. To prevent stifling private efforts, central governments might avoid using this as an
opportunity to influence the allocation of funds.
Available information indicates that trafficking of women and children is pervasive in
many African countries, due to an inadequate legal framework, a lack of resources to enforce the
law, and perhaps social attitudes and practices that condone trafficking. Expanding information
on the victims of trafficking can play a useful role in gathering support for reforms. For
example, there is some evidence that the terrible reputation earned by Nigeria has spurred efforts
to restrain trafficking in that country. If this goal is adopted, it would be important to consult
with the NGOs that are active in fighting trafficking.
Surveys of the channels used to send remittances, the reasons for choosing them, and the
costs involved could help focus government efforts to improve the remittance market. At the
same time, interviews with policy makers and companies involved in the market may help to
clarify the tradeoffs facing efforts to expand remittance services. For example, there is some
evidence that financial regulation in West Africa is an impediment to developing formal
remittance channels that serve poor areas. Perhaps there is scope for modifying the regulatory
framework to permit more small scale remittance agencies, without eroding safeguards against
fraud and financial irresponsibility. In South Africa, it may be possible to ease informational
requirements governing remittance providers that effectively shut low-income workers out of the
formal market.
The literature on migration also provides some guidance on areas of inquiry that could be
less productive. High-skilled emigration is likely an important issue for some countries,
particularly for health services, but it is not clear what contribution surveys or interviews would
make. Moreover, we need a better understanding of the causes and consequences of high-skilled
emigration, and the influence of other reasons for deterioration in services, to formulate useful
policies. It is not likely that the Africa project would effectively address this issue. The role of
migration in spreading disease, in particular HIV/AIDS, has been well documented. Survey
information could conceivably assist in identifying populations that require services, but in
general the role of the project in this field is unclear. Analyses of how remittances are spent
don’t appear to generate useful information, and require intensive analysis to be done correctly.
31
Overall, at this stage it may be more productive to focus on modifying policy
interventions that currently reduce welfare and selecting small programs that could improve the
benefits of migration, rather than devising broad, migration-friendly policy frameworks.
Particularly in the poor countries of West and East Africa, governments are not in a position to
implement comprehensive policies that would make a noticeable contribution to improving the
development impact of migration. However, it might be possible to make targeted changes that
help in specific areas. With migration in Africa, the best may be the enemy of the good.
32
Annex on Individual Countries
Introduction
This annex provides information on migration issues that is not appropriate for the
discussion of common issues provided above. Some of it simply emphasizes points made in the
main body of the paper, other parts reflect characteristics of migration that are not sufficiently
common among the 10 countries to mention above. Thus each country discussion is more a
collection of observations than a coherent story.
The annex also provides some additional data on migration in the sample countries,
based on work by the United Nations Population Division and the World Bank Development
Prospects Group. The information provided on a cross-country basis includes, for time series,
migration stocks in destination countries and remittance receipts in origin countries, and for
2005, the main sources of immigration and destinations of emigration. These latter data are the
result of efforts to construct a comprehensive matrix of migration stocks, based initially on the
UN Population Division data on the composition of immigrants by origin countries, along with
additional information, and in some cases estimations based on reasonable inferences.
1.
Burkina Faso
Immigration to Burkina Faso has increased rapidly over time, from less than 2 percent of
the population in 1970 to almost 6 percent in 2005 (table A1.1), despite population growth of 2.6
percent per year over this period. The accuracy of these data may be open to question, however,
as many migrants to and from Burkina Faso do not cross official border posts (Pizarro 2006). At
the same time, remittances received from Burkinabe abroad have declined sharply, from a peak
of 16 percent of GDP in 1980 to just under 2 percent by 2005. In part the sharp fall in dollar
value of the early 1990s reflects the devaluation of the CFA franc. Burkina Faso’s emigrants are
overwhelmingly in Cote d’Ivoire (table A1.2), and the economic crisis in that country has no
doubt reduced remittances due to lower earnings and greater numbers of returnees. There may
also be issues about the quality of the remittance data. Remittance receipts are exactly $50
million in every year from 2001-2005, raising questions about whether these reflect actual
reports or guesstimates.
Table A1.1 Migration and remittances over time
Immigrants
thousand persons
percent of population
Remittances
million US$
percent of GDP
..
..
1970
1975
1980
1985
1990
1995
2000
2005
89
1.7
106
1.8
170
2.6
273
3.7
345
4.0
464
4.7
573
5.1
773
5.8
48
10.5
150
16.0
126
6.5
140
8.7
80
2.6
67
2.7
50
1.9
Sources: United Nations Population Division, World Bank Remittances and Migration website
33
Table A1.2 Distribution of migration, 2005
(percent of total)
Benin
Cote d'Ivoire
Ghana
Mali
Niger
Other identified
Unidentified
Immigrants
6.5
3.1
8.3
39.0
12.3
0.0
30.8
Cote d'Ivoire
Niger
Nigeria
France
Italy
Other identified
Unidentified
Emigrants
88.4
1.0
0.6
0.5
0.4
0.6
8.5
Source: World Bank Remittances and Migration website
Sociological and cultural factors have a role in determining migration patterns in Burkina
Faso. The Fulani population in the Sahel region of northern Burkina Faso supplies a large
proportion of the country’s emigrants. According to Hampshire and Randal (1999), emigration
is to some extent determined by ethnic identity, with some of the Fulani subgroups unlikely to
emigrate unless desperate. Konseiga (2005a) notes that some migration results from marriages,
divorces, widowhoods. Moreover, the dependence on ones parents traditionally lasts into the late
30s, so many young people use emigration as a means of escaping such constraints, and as an
initiation process and adventure. Konseiga (2005b) emphasizes that pastoralists are unlikely to
migrate because, unlike farmers, they cannot safely leave their flock behind.
The main body of the report provides data on high-skilled emigration to OECD countries.
But Thiombiano (2000) cites a large high-skilled emigration from Burkina Faso to Cote d’Ivoire,
and that in total 74 percent of those educated at college level have left the country.
2.
Ethiopia
Immigration to Ethiopia increased sharply, both in terms of the number of people and
relative to the population, from 1970 to 1990, but then dropped by over half in numbers and from
2.3 percent to 0.7 percent of the population, reflecting the severe economic problems and
political violence afflicting the country (table A2.1). At the same time, remittances have
mushroomed from $5 million in 1990 to $134 million in 2005, as the Ethiopian diaspora has
increased significantly (table A2.2). More than half of the total 446,000 Ethiopian emigrants are
in high-income countries, including the roughly 79,000 Falasha (Ethiopian Jews) who left for
Israel. But according to the official data, the largest émigré Ethiopian community is in the
Sudan, including former refugees who never returned home. Over four-fifths of immigrants in
Ethiopia were either born in Eritrea when the two countries were united, or have escaped the
famine and wars in the Sudan.
34
Table A2.1 Migration and remittances over time
Immigrants
thousand persons
percent of population
Remittances
million US$
percent of GDP
..
..
1970
1975
1980
1985
1990
1995
2000
2005
395
1.3
392
1.1
404
1.1
584
1.3
1,155
2.3
795
1.3
662
1.0
555
0.7
12
14
0.1
5
0.0
27
0.4
53
0.7
134
1.2
..
..
..
Sources: United Nations Population Division, World Bank Remittances and Migration website
Distribution of migration, 2005
(percent of total)
Immigrants
Djibouti
Eritrea
Kenya
Somalia
Sudan
Other identified
Unidentified
Emigrants
0.7
53.0
0.1
29.4
1.7
0.0
15.1
Germany
Israel
Saudi Arabia
Sudan
United States
Other identified
Unidentified
4.5
17.8
5.9
28.9
17.9
16.5
8.5
Source: World Bank Remittances and Migration website
Exploitation of Ethiopian women emigrants is a serious issue. Many women use Moslem
pilgrimages as a pretext to go to Saudi Arabia, drawn by false and misleading information on job
opportunities and conditions (Kebede 2001). Some are required to sign domestic service
contracts they cannot read after payment of part of the fee, with clauses such as prohibitions on
leaving the house without permission and large penalties if they quit their jobs. These
requirements make them extremely vulnerable to abuse from their employers. Kebede (2001)
reports many cases of rape or attempted rape by male employers and women left in jail with no
legal support. Even with government-provided employment services, there is no pre-departure
training or information provided on living conditions in destination countries. In response to this
situation, a private employment agency was established by Ethiopian women who used to live in
Lebanon to provide women with skills training and orientation.
3.
Ghana
Migration in Ghana is a longstanding tradition. Virtually all ethnic groups claim to have
emigrated from somewhere else (Anarfi and others 2003). Ghana shifted from a country of
immigration during the colonial era and up to the late 1960s to a substantial source of emigrants
as economic troubles deepened in the 1970s and 1980s. With the return of economic progress
from the mid-1990s, however, immigration to Ghana has increased significantly, from 4.6
percent of the population in 1990 to 7.5 percent in 2005 (Table A3.1). Ghana does not provide
data on the composition of its immigrant population, so the only data available on Ghanaian
35
migration is from partner countries. Emigration from Ghana is largely to other West African
countries (particularly Cote d’Ivoire and Nigeria—table A3.2), which account for two-thirds of
total emigrants. Emigration from Ghana is largely dominated by men, with the exception of the
substantial flows to Cote d’Ivoire (Anarfi and others 2003), and often takes place without
appropriate exit documents (Adepoju 2005). Forced migration has occurred both to and from
Ghana. It is one of several countries of asylum for fleeing civil wars in Liberia, Sierra Leone and
Cote d’Ivoire (Anarfi and others 2003), Ghana expelled foreigners in 1969, and Ghanaians were
expelled from Nigeria in 1983, although many returned in the late 1980s (Higazi 2005). A
growing share of the poor emigrated from Ghana over the 1990s, owing to declining
employment opportunities in the urban sector and declines in the terms of trade of cash crops
(Litchfield and Waddington 2003). The probability of migration was positively related to the
level of education, previous migration experience, access to irrigation, household size, and the
availability of networks, and negatively related to the household dependency ratio and the
potential for off-farm employment (Tsegai 2004)
Table A3.1 Migration and remittances over time
Immigrants
million persons
percent of population
Remittances
million US$
percent of GDP
..
..
1970
1975
1980
1985
1990
1995
2000
2005
352
3.9
385
3.8
421
3.7
494
3.7
717
4.6
1038
5.9
1505
7.6
1669
7.5
1
0.0
4
0.1
6
0.1
17
0.3
32
0.6
99
0.9
..
..
Sources: United Nations Population Division, World Bank Remittances and Migration website
Despite sharp increases in remittances since 1990, they remain less than 1 percent of
GDP (table A3.2). Remittances from outside Africa are estimated at 37 percent of the total
(including internal remittances), while emigrants outside Africa are only 12 percent of Ghanaians
sending remittances (Mazzucato and others 2005). It can be difficult to distinguish between
internal and external remittances, as transfers from an internal source may have been financed by
transfers from abroad. Ghanaians who worked while abroad, had better jobs, stayed abroad
longer, and maintained contact with their families are more likely to send money home and sent
larger amounts (Black and others 2003a). Returned migrants also benefited from the experience:
less-skilled emigrants showed a significant improvement in occupational level, with those who
left when they were younger showing the most improvement.
Addison (2004) reports that the cost of remittance transfers is falling due to increased
competition. A large proportion of remittances come through foreign exchange bureaus, which
are more prevalent than either banks or formal money transfer operators (only about 5 percent of
Ghanaians have a bank account). Trust is an important element of these transfers, as they
typically involve the exchange of a large number of bills that take time to count. Remittances
also are provided in the form of goods, largely consumer durables.
Statistics on emigration of health professionals are disturbing: the share of doctors
emigrating in the total trained in a given year rose from 60 percent in 1995 to 94 percent in 2002
36
(Anarfi and others 2003). In turn, Ghana is a significant importer of Cuban doctors (Bach 2003).
The share of nurses and midwives leaving is reported as averaging 20 percent of those trained
from 1995-2002, although it is very difficult to obtain accurate data on nurses emigrating
(Buchan and others 2003). However, the number of nurses emigrating dropped by more than half
in 2003, in part due to the United Kingdom’s prohibition of recruiting of African health
professionals by the National Health Service (Quartey 2006). The raw numbers may understate
the impact of emigration on nursing in Ghana, as 95 percent of those leaving are highly
experienced.
Table A3.2 Distribution of migration, 2005
(percent of total)
Burkina Faso
Cote d'Ivoire
Germany
Nigeria
Togo
United States
Other identified
Unidentified
Emigrants
7.1
34.2
3.6
17.7
4.3
8.2
16.5
8.5
Note: No information is available on the distribution of immigrants
Source: World Bank Remittances and Migration website
4.
Lesotho
Emigration from Lesotho is conditioned by its being surrounded by South Africa, by its
dependence on South Africa for income, and by the strong ethnic ties between Basothos in the
two countries. Historically, Lesotho was established by the Basotho king and eventually
sustained, albeit in much reduced form, by British intervention against the Boers. As a result,
many Basotho pay only marginal attention to the border between the two countries, and some
writers emphasize that the degree of distinction between the two countries is not always clear.
Coplan (2001) concludes that South Africa lacks the resources, political will, or even moral
authority to prevent or reverse immigration from Lesotho. A survey found that travel to South
Africa was a way of life for the overwhelming majority of Basotho, and that most respondents
saw the border as unnecessary (Gay 2000). Eighty percent of Lesotho’s emigrants go to South
Africa (table A4.2), propelled by small farm size and marginal ecological conditions (Afolayan
2001), the much greater economic opportunities in South Africa (particularly the need for labor
in the mines), and since the late 1990s, political instability and slow economic growth.
Emigration to South Africa is thus a dominant feature of Lesotho’s economic life. A 2003
survey found that 37 percent of those interviewed reported a family member working in South
Africa, 26 percent a family member permanently settled there, and 18 percent that admitted
holding South African identification cards (Cobbe 2004). Immigration to Lesotho is minimal,
about 0.3 percent of the population, owing to the lack of economic opportunities.
37
Table A4.1 Migration and remittances over time
Immigrants
thousand persons
percent of population
Remittances
million US$
percent of GDP
1970
1975
1980
1985
1990
1995
2000
2005
4
0.3
4
0.3
9
0.7
16
1.1
7
0.4
5
0.3
5
0.3
6
0.3
122
81.6
263
61.0
224
77.3
428
69.5
411
44.1
252
29.3
355
24.4
..
Sources: United Nations Population Division, World Bank Remittances and Migration website
Remittances fell dramatically from 1990 to 2000, and despite some recovery by 2005
remained at 24 percent of GDP, which is very large compared to most developing countries but
less than one-third of the peak level in 1985 (table A4.1). Since the mid-1990s, the demand for
immigrant labor in South Africa’s mines has fallen sharply, and South African policies have
remained relatively hostile to immigration (see below). Economic growth fell to 2.2 percent per
year from the completion of major infrastructure projects in 1998 to 2005, compared with 4.7
percent in the decade before.
An interesting aspect of Lesotho emigration has been the deferred pay scheme
established in 1974 to control remittances by miners in South Africa. In principle, mining
companies transfer 30 percent of their wage bill to a collective account at Lesotho Bank
(Sparreboom and Sparreboom-Burger 1996). At the end of their one-year contract, mine
workers receive a deferred pay certificate which they can use to withdraw their funds in Lesotho
(they are also permitted to make two emergency withdrawals during the year). The fund’s
administrative responsibilities include enforcing contributions by the mining companies. The
system was not very popular with mineworkers: participation in voluntary schemes for mine
workers from Botswana and Mozambique is considerably lower than Lesotho’s compulsory
scheme, and calls for its abolition figured in the wage negotiations in 1994. However, the
government of Lesotho benefits because it has use of the funds until the miners are paid at the
end of each year (Sander and Maimbo 2003).
Table A4.2 Distribution of migration, 2005
(percent of total)
Immigrants
Botswana
South Africa
Tanzania
Zambia
Zimbabwe
Other identified
Unidentified
Emigrants
1.4
50.9
1.4
1.2
1.5
2.0
41.6
Mozambique
South Africa
Tanzania
United Kingdom
United States
Other identified
Unidentified
10.0
80.5
0.3
0.1
0.1
0.3
8.5
Source: World Bank Remittances and Migration website
38
5.
Mali
Immigration to Mali, a poor, landlocked country with a per capita income of less than
$400, remains minimal. Remittances declined from 1995 to 2000 in dollar terms, but then
doubled over the next five years (table 5.1). During this period, the CFA, which is tied to the
euro, appreciated strongly against the dollar, and the more than doubling of the dollar amount of
remittances from 2000 to 2005 reflects an increase of just under 60 percent in terms of CFA
(about 90 percent of Malian emigrants go to countries whose currency is tied to the euro). Even
so, the rise in remittances seems odd, given that about half of Malian emigrants go to either Cote
d’Ivoire or Nigeria, countries which have experienced economic recession and political
instability over the past few years. As the same dollar figure is given for 2002-2005, it is
possible that reporters could have assumed the same nominal value in the absence of
information.
Table A5.1 Migration and remittances over time
Immigrants
thousand persons
percent of population
Remittances
million US$
percent of GDP
..
..
1970
1975
1980
1985
1990
1995
2000
2005
95
1.7
82
1.3
71
1.0
61
0.8
60
0.7
63
0.6
48
0.4
46
0.3
23
2.8
59
3.3
67
5.1
107
4.4
112
4.5
73
3.0
155
3.0
Sources: United Nations Population Division, World Bank Remittances and Migration website
Destinations for Malian emigrants vary considerably. International migrants may go to
France, to the peanut plantations of Senegal and Gambia, the cotton and cocoa plantations of
Cote d’Ivoire, and to neighboring regions seeking seasonal availability of grass for herds
(Findley 2004). Many migrants travel to neighboring countries for relatively short (less than six
months) periods, while emigration to Europe tends to be for longer periods. Short-term
circulation to neighboring countries may rise during a drought, but emigration to Europe may fall
due to a lack of resources. Hence surveys show that the total amount of emigration from
drought-stricken regions does not change greatly during drought. And higher income is
associated with higher rates of long-term emigration, but lower rates of short-term (Russell and
others 1990).
39
Table A5.2 Distribution of migration, 2005
(percent of total)
Emigrants
Burkina Faso
Cote d'Ivoire
France
Gabon
Niger
Nigeria
Other identified
Unidentified
24.8
40.7
3.7
2.7
2.9
9.5
7.2
8.5
Note: No information is available on the distribution of immigration
Source: World Bank Remittances and Migration website
The government cooperates with French efforts to return unauthorized emigrants in
France, and works with international organizations to attract educated Malians (Martin and
others 2002). The government has established a ministry devoted to the Malian diaspora which
(for example) provides information on the availability and costs of money transfer services and
maintains an inventory of skills that can contribute to Mali’s educational system and small
business development. The government also provides information on foreign employment
opportunities (Sander and Maimbo 2003). These efforts have helped mobilize support from
France, including a 2.6 million euro project to cofinance local development projects, support
reintegration of migrants, and mobilize Mali’s scientific diaspora.
Mali also receives considerable resources from the diaspora for public projects. A mid1990s survey found that 70 percent of emigrants to France from the Kayes region (the main
source of Malian émigrés) are active members of village associations, which play an important
role in development (AFFORD 2000). Over 10 years, village associations financed 146 projects
for a total of 19.4 million French francs. Raunet (2005) reports that 22 percent of total remittance
receipts comes from groups financing local community projects. Survey respondents also felt
that return migration has encouraged democratic reforms, although concerns have been
expressed about the importation of ‘decadent’ western values, particularly concerning the role of
women and children. For example, one European official noted Malian concern that traditional
aspects of local culture, such as polygamy and genital mutilation, were illegal in European
destination countries (Martin and others 2002).
Remittances equal about 3 percent of GDP for the country as a whole. However,
remittances make up a very significant share of income in the Kayes region, the main source of
Malian emigrants. Ammarrasi (2005) reports an early 1990s survey showing that remittances
account for 80 percent of household resources in areas of the Senegal River valley. Azam and
Gubert (2004) find that large remittance receipts in Kayes have lowered incentives to work.
Emigration of some household members is viewed as insurance against a shortfall in
consumption during bad times. However, emigrants cannot easily monitor work effort, leading
to some moral hazard. Households with significant access to remittances show worse
agricultural performance than households without members abroad, even though remittances
have contributed to the adoption of improved agricultural technology.
40
6.
Mauritius
Of all the sample countries, Mauritius is the least affected by migration. Immigration is
less than 2 percent of the population (table A6.1). One of the few middle-income countries in
the Sub-Saharan Africa region, it is protected by the Indian Ocean from irregular migration. The
government effectively controls entry, and the number of visa overstays is limited (Hein 2004).
China and India account for about half of the stock of immigrants (table A6.2). Government
policy is not to accept immigrants for unskilled work, and entry permits generally are not issued
for domestic service. However, some of the ‘skilled workers’ permitted entry have job titles
such as knitting operators, painters, metal sorters, fish cutters and apprentices (Hein 2004). The
government also follows a strict policy of rotating guest workers to avoid the problems with
guest worker programs in Europe.
Remittances have grown over the past decade in dollar terms, remaining between 3 and 4
percent of GDP. Most emigration is to high-income countries (table A6.2). Easier entry rules
for highly-educated workers have encouraged substantial emigration from Mauritius in recent
years, and 56 percent of its college-educated citizens live abroad. Mauritius has a greater
percentage of managers that cite the skills and education of workforce as major constraint on
their operations than in South Africa, Malaysia, Indonesia and India (EU 2006). The government
is increasing efforts to attract more investors and skilled professionals by broadening eligibility
for work permits and residence permits.
Table A6.1 Migration and remittances over time
Immigrants
thousand persons
percent of population
Remittances
million US$
percent of GDP
..
..
1970
1975
1980
1985
1990
1995
2000
2005
11
1.3
10
1.1
10
1.0
9
0.9
9
0.8
12
1.0
16
1.3
21
1.7
132
3.5
177
4.0
215
3.3
..
..
..
..
..
..
..
..
Sources: United Nations Population Division, World Bank Remittances and Migration website
Table A6.2 Distribution of migration, 2005
(percent of total)
Immigrants
China
France
India
South Africa
United Kingdom
Other identified
Unidentified
Emigrants
37.0
10.7
15.9
5.6
7.2
12.5
11.1
Australia
Canada
France
Italy
United Kingdom
Other identified
Unidentified
14.3
6.0
27.8
9.6
25.2
8.6
8.5
Source: World Bank Remittances and Migration website
41
7.
Nigeria
During the oil boom from the early 1970s to the early 1980s, Nigeria became a
significant destination for immigrants from other West African countries, and immigration
reached almost 2 percent of the population, according to official statistics (table A7.1). The
extent to which immigration during this period (which was heavily weighted towards unskilled
workers from neighboring countries—Black 2004) was accurately measured is unclear. But the
severe downturn in the economy with the decline in oil prices led to the expulsion of many
foreigners in 1983, with the government claiming that many were engaged in street begging and
prostitution, or employed in business and trade without permission (Afolayan 1988). Some 1.3
unauthorized migrants left the country in 1983, leading to camps along the border as some of the
other West African states closed their borders against the influx of refugees. Fully 80 percent of
immigrants in 2005 came from other West African countries. While emigration to West Africa
also is significant, North America and Europe accounted for more than 40 percent of Nigeria’s
emigrants in 2005.
Table A7.1 Migration and remittances over time
Immigrants
thousand persons
percent of population
Remittances
million US$
percent of GDP
..
..
1970
1975
1980
1985
1990
1995
2000
2005
158
0.3
704
1.2
1315
1.9
348
0.4
447
0.5
582
0.6
751
0.6
971
0.7
22
0.0
10
0.0
10
0.0
804
2.9
1,392
3.3
2,273
2.3
..
..
Sources: United Nations Population Division, World Bank Remittances and Migration website
Some 11 percent of highly-educated Nigerians live in OECD countries, in part due to
limited opportunities for tertiary education in Nigeria (Black 2004). The government has made
efforts to generate benefits from the many highly-educated Nigerians residing abroad. An office
was established for diaspora activities and a database on the skills of emigrants was set up
(Ammarrasi 2005). AFFORD (2000) notes the strength of Nigerian hometown associations, in
part due to the trauma of the civil war, which underlined the importance of having a secure home
base in times of trouble. Nigerians abroad also play an important role in local politics. For
example, the Association of Nigerians Abroad has campaigned for democracy and against
human rights violations, while the separatist movement enjoys considerable support among the
diaspora (Carling 2006).
42
Table A7.2 Distribution of migration, 2005
(percent of total)
Immigrants
Benin
Chad
Ghana
Mali
Niger
Togo
Other identified
Unidentified
Emigrants
21.2
2.4
16.5
11.8
7.8
10.3
10.7
19.3
Benin
Cameroon
Chad
Niger
United Kingdom
United States
Other identified
Unidentified
6.2
9.5
15.3
5.1
11.7
18.6
25.0
8.5
Source: World Bank Remittances and Migration website
Weak enforcement of laws and ease in counterfeiting documents have made Nigeria a
center for the trafficking of women and children. The government sponsors information
campaigns on child labor and has beefed up police and immigration units dedicated to fighting
trafficking (Black 2004). Nevertheless, trafficking remains a serious problem, and creative
approaches to circumventing enforcement in both Nigeria and Europe are evolving. For
example, under-aged girls seeking asylum in the Netherlands are placed with foster parents until
their cases are resolved. Many Nigerian girls requested asylum and subsequently disappeared
from their foster homes. It turned out that the asylum system was being used by traffickers to
smuggle girls to the Netherlands (Carling 2006).
8.
Senegal
Immigration to Senegal remained relatively stable from 1990 to 2005, although rapid
population growth led to a decline in the share of immigrants in the population to below 3
percent (table A8.1). Unlike several other West African countries, remittances increased
strongly over this period, more than doubling as a share of GDP. This is probably due to the fact
that about 45 percent of emigrants are to high-income countries, and emigration to crisis-plagued
Cote d’Ivoire is more limited from Senegal than from Burkina Faso and Mali. Nevertheless, the
official data listing no emigrants from Senegal to Cote d’Ivoire cannot be correct; Fall (2003)
reports that there were 100,000 Senegalese in Cote d’Ivoire in 1998. While the total of
emigrants (estimated on the basis of partner country reports) of 463,000 is not huge relative to
the population of 11.7 million, some areas of Senegal have a large concentration of emigrants.
For example, Hearing and van der Erf (2001) note villages where one in every two households
have migrants abroad.
43
Table A8.1 Migration and remittances over time
Immigrants
thousand persons
percent of population
Remittances
million US$
percent of GDP
1970
1975
1980
1985
1990
1995
2000
2005
176
3.8
155
2.9
119
2.0
170
2.5
293
3.7
320
3.5
297
2.9
326
2.8
26
1.4
77
2.6
79
3.1
142
2.5
146
3.3
233
5.3
511
6.1
..
..
Sources: United Nations Population Division, World Bank Remittances and Migration website
Table 8.2 Distribution of migration, 2005
(percent of total)
Immigrants
Cape Verde
France
Guinea
Guinea-Bissau
Mali
Mauritania
Other identified
Unidentified
Emigrants
4.4
8.2
38.5
11.5
8.3
14.1
5.6
9.4
France
Gabon
Gambia, The
Italy
Mauritania
United States
Other identified
Unidentified
19.5
4.0
26.6
15.3
9.3
2.6
14.2
8.5
Source: World Bank Remittances and Migration website
Hearing and van der Erf (2001) report data from a survey on information available to
Senegalese emigrants. The majority of those leaving had some information on the intended
country of destination, but very few knew anything on admissions procedures, indicating the
high level of irregular migration. Migrants made only limited use of agencies as a source of
information. Sander and Barro (2003) note the importance of ethnic and religious networks that
facilitate departure, arrival, and integration into destination countries.
Senegal has taken steps to promote the welfare of emigrants. Agreements in the 1990s
with France and Gabon provided that salaried emigrants would be entitled to social benefit
provided by destination countries. The government also has encouraged return through
cooperating with France, although programs in the 1980s showed poor results in terms of the
sustainability of businesses and loan repayments (Diatta and Mbow 1999), while efforts in the
1990s had few participants (Raunet 2005).
The government also promotes the involvement of the diaspora in economic and social
development. Ammarrasi (2005) notes the considerable contribution of emigrants to local
communities. More than a third of émigrés from Senegal are members of a village association,
and there is evidence that emigrants’ use of telecommunications technology has contributed to
social transformation in rural areas. Russell and others (1990) describe the practice of setting up
44
‘secondary villages’ among emigrant groups that take responsibility for building and furnishing
schools. It is possible that these efforts, or emigration in general, have contributed to a
breakdown of traditional authority. For example, Guilmoto (1998) notes that the influence of
village, caste, lineage and extended family is seen to be eroded by emigration out of rural areas,
as well as by state intervention.
Migration plays an important role in spreading the AIDS epidemic in Senegal. A study
of 11 villages around Matam in the north found that 27 percent of returning male migrants were
infected with HIV, compared with less than 1 percent of males who had not migrated (Thiam and
others 2003). And 40 percent of migrants to African countries with high rates of infection
reported having risky sexual behavior, such as sexual contact with prostitutes or casual partners.
Nevertheless, less than 2 percent of the adult population is HIV-positive, compared to the
average of nearly 9 percent for Sub-Saharan Africa. Success in restraining the epidemic is
attributed to early government interventions plus cultural and religious constraints on sexual
activity.
Formal means of transferring money to Senegal include the postal system, which
involves long delays; bank transfers between accounts, which are reliable but expensive; and
money transfer operators, which are even faster but very costly (Sander and Barro 2003). There
is some evidence of improvements in the efficiency of money transfers. For example, the Kara
International Exchange allows migrants to deposit money with a business in New York. The
migrant is given a number which she communicates to the beneficiary in Senegal, who can
obtain the funds at a local business (Addy and others 2003). This arrangement provides for the
bundling of many small transactions so that transfers can be effected at lower rates. Growing
amounts of remittances are being transferred through telephone or fax (Tall 2003). Microfinance
institutions are cooperating with banks to offer transfer services, and banks are compensating for
their limited networks (only 6 percent of the population have bank accounts) by offering points
of service at telecenters and cybercafes (Sander and Barro 2003). Banks also are forming
alliances with Western Union and Moneygram to boost their participation in the money transfer
business.
9.
South Africa
Probably more has been written on migration to South Africa than any other African
country, yet intense controversies remain over the number and impact of immigrants. Official
data report 1.1 million immigrants (table A9.1), but of African countries, immigrants are
reported from only Botswana, Lesotho, Malawi, Mozambique, Swaziland, and Zimbabwe (some
of these communities originated in movements of tribal groups in the 19th century) (Solomon
1996). Estimates of irregular immigration vary from 500,000 to 8 million (Crush and Peberdy
2003, Solomon 1996). The techniques used to estimate the number of irregular migrants,
essentially extrapolations from repatriations and visa overstays, are clearly imprecise. Border
controls are ineffective due to lack of resources (Minnaar 2001 cites the use of horse and foot
patrols to patrol parts of the border owing to petrol shortages). There is some evidence that
much irregular immigration is not intended to be permanent. A survey in Swaziland found that
56 percent of respondents had a strong or moderate desire to go to South Africa for a period of
45
up to 2 years, but only 32 percent expressed a strong or moderate desire to live in South Africa
(Simelane and Crush 2004). Shaw (2001) claims that West African criminal networks, mainly
Nigerian, have contributed to the growth of organized crime in South Africa.
Table A9.1 Migration and remittances over time
1970
1975
1980
1985
1990
1995
2000
2005
Immigrants
thousand persons
percent of population
961
4.2
962
3.7
983
3.4
1815
5.5
1225
3.3
1098
2.6
1022
2.2
1106
2.3
Remittances
million US$
percent of GDP
15
0.1
36
0.1
67
0.1
39
0.1
136
0.1
105
0.1
344
0.3
658
0.3
Sources: United Nations Population Division, World Bank Remittances and Migration website
The decline in legal immigration since the early 1990s reflected a more restrictive stance
on immigration since the ending of apartheid (Crush and Peberdy 2003), evidenced by the
greater difficulties employers faced in obtaining permits for foreign workers. Crush and
McDonald (2001) claim that beyond some limited amnesties for immigrants that had been in the
country for extended periods of time, the post-apartheid government has shown little appetite for
immigration. Klaaren and Ramji (2001) find that South Africa’s policing policy towards
immigration changed little after the demise of apartheid. Landau (2005) claims that nativism has
helped create ‘zones of exception’ where South Africa’s normal legal provisions are suspended
so that police can use illegal methods against, and exploit money from, migrants. And Danso
and McDonald (2001) note that the coverage of immigration in the South African press has been
largely anti-immigrant. This in part reflects fear of competition on the part of poor workers.
Interviews suggest that African, non-national small scale entrepreneurs have more access to
capital and are more educated than their South African counterparts (Peberdy and Rogerson
2000). And some researchers claim that employers discriminate in favor of immigrants, perhaps
due to the willingness of irregular migrants to work for low wages in poor conditions, owing to
their greater vulnerability than native workers (Zuberi and Sibanda 2004). Some softening of the
stance towards immigrants can be found in the immigration law adopted in 2003. However, the
governments still seeks to control irregular migrants through deportations rather than pressure on
employers to comply with the law (Human Rights Watch 2006).
Controversy surrounds the reasons for, extent, and impact of high-skilled emigration
from South Africa. The number of emigrants is significantly underestimated, as official records
only count those who declare their intention to emigrate when leaving the country (Bhorat and
others 2002). Calculations of flow data on highly-educated emigrants in major receiving
countries find that professional emigration is more than 3 times official figures. The net loss of
skills is also due to a significant decline in skilled immigration. Wocke and Klein (2002) cite an
ILO study finding that a chronic shortage of skilled labor has impaired economic growth and
hindered development of labor-intensive sectors. A survey in 1998 of highly-educated workers
found that 65 percent of skilled whites felt their standard of living had deteriorated since
apartheid ended in 1994, while 65 percent of blacks felt theirs had improved. A 2002 survey of
46
725 skilled workers found that 70 percent had thought of emigrating, but only 7 percent intended
to leave within the next six months (OECD 2003). The probability of leaving was found to be
very high for 2 percent of those surveyed, and high for 10 percent.
The scarcity of health workers appears to be concentrated in the public sector. The
Department of Health estimates more than 4 thousand unfilled vacancies for physicians and
almost 33 thousand unfilled vacancies for nurses, representing a little over a quarter of the total
number of vacancies for these two groups (OECD 2003). By contrast, there appears to be a
surplus of nurses in the private sector. Emigration of nurses appears to have risen over time,
supported by the growth of overseas nurse associations and other support networks, particularly
in the United Kingdom (Bach 2003). Overall, however, emigration is not the fundamental
reason for shortages of health professionals, as the 32,000 nurse vacancies in the public health
sector are matched by 35,000 registered nurses in South Africa that are inactive or unemployed.
The major reasons given for emigration of health professionals were crime, affirmative action,
the deteriorating state of public education, uncertainties about the future, the fragility of the
economy, the transferability of South African qualifications to OECD countries, integration into
a knowledge-based global economy, and the activity of foreign recruitment agencies.
Table A9.2 Distribution of migration, 2005
(percent of total)
Immigrants
Botswana
Lesotho
Mozambique
Swaziland
Zimbabwe
Malawi
Other identified
Unidentified
Emigrants
2.2
18.8
24.4
7.3
46.1
1.0
0.2
0.0
Australia
Canada
Mozambique
Namibia
United Kingdom
United States
Other identified
Unidentified
11.2
5.6
19.6
5.1
22.0
10.6
17.3
8.5
Source: World Bank Remittances and Migration website
South Africa’s remittance transfer system is in the process of change to cope with
shortfalls in the formal system of transfers. Postal services can be unreliable and slow (Gupta
and others 2006), and many post offices cannot handle money orders owing to the costs and risks
involved (Sander and Maimbo 2003). The speed and security of postal transfers depends in part
on the post office in the receiving country and its ties to South Africa (CGAP 2003). Many rural
bank branches closed during the 1990s owing to the economic decline, a rise in violent robberies,
and the spread of internet banking among higher income groups which undercut the rationale of
local banking structure. Where available, ATMs are an increasingly popular means of getting
around restricted banking hours, although they recently have become the target of bombings to
obtain cash (Washington Post 2007).
The formal system is innovating to overcome some of these disadvantages. The post
office has introduced a new PIN money order which provides for on-line, immediate funds
47
transfer (CGAP 2003), although this appears to be limited to within-country transfers.
Nevertheless, informal transfer agents apparently remain strongly competitive. One survey
reported that a 250 rand transfer abroad cost 22 rand if done by a friend, 40 by postal order, 50
by a taxi driver, 80 to 125 by a money transfer operator, and 160 by a commercial bank.
Moreover, regulatory provisions tend to exclude irregular migrants and, by adding to transactions
costs, low-income migrants, from formal systems.
10.
Uganda
Economic disintegration and conflict have severely depressed Ugandan living standards,
and immigration has fallen precipitously since more prosperous times (table A10.1). The
principal destination countries for Ugandan émigrés are the United Kingdom and Tanzania, with
emigrants to the latter often escaping civil conflict (table A10.2). By end-2005 there were
38,000 Ugandan refugees and asylum-seekers.
Table A10.1 Migration and remittances over time
Immigrants
thousand persons
percent of population
Remittances
million US$
percent of GDP
..
..
1970
1975
1980
1985
1990
1995
2000
2005
953
10.3
777
7.2
678
5.4
634
4.3
550
3.1
610
2.9
529
2.2
518
1.8
238
4.0
642
7.4
..
..
..
..
..
..
..
..
..
..
Sources: United Nations Population Division, World Bank Remittances and Migration website
Official data on remittances show a sharp rise from 2000 to 2005, but historical data are
not available (Table A10.1). Sander and Maimbo (2003) attribute the rise in remittances to the
liberalization of financial markets, particularly the allowing residents to open domestic accounts
denominated in foreign currency. Postal orders are the cheapest means of transferring money
(6,000 shillings to send 100,000), and commercial banks and money transfer operators the most
expensive (17,000 to 35,000 shillings to send 100,000).
48
Table A10.2 Distribution of migration, 2005
(percent of total)
Immigrants
Burundi
Congo, Dem Rep.
Kenya
Rwanda
sudan
Tanzania
Other identified
Unidentified
Emigrants
15.7
13.2
6.4
19.2
29.6
11.1
0.7
4.2
Canada
Rwanda
Tanzania
United Kingdom
United States
Sweden
Other identified
Unidentified
7.7
3.5
23.4
39.7
8.7
1.6
7.0
8.5
Source: World Bank Remittances and Migration website
The problems besetting Uganda over the past few decades have led to a large exodus of
skilled workers. It is estimated that Uganda lost more than half of its professional and technical
manpower during the rule of Idi Amin from 1971-79, with continuing outflows over the
subsequent decades (Black and others 2004). But at the same time, Russell and others (1990)
claim that education expanded faster than the absorptive capacity of the economy, so that trained
workers were unable to find jobs at home. In the late 1980s, Uganda attempted to limit the
emigration of professionals and civil servants by foreign exchange restrictions and the
requirement of clearance for foreign travel, but these had disappeared by 2000. A number of
Ugandan diaspora groups in both North America and Europe provide assistance with
development or promote conflict resolution in the home country
(www.buganda.com/diaspora.htm).
49
REFERENCES
Adams, Richard. 2006. “Remittances in Poverty in Ghana.” World Bank Policy Research
Working Paper 3838. February. www.worldbank.org.
Adams, Richard and John Page. 2003. “International Migration, Remittances, and Poverty in
Developing Countries. World Bank Policy Research Working Paper No. 3179.
Addison, E. K. Y. 2004. “The Macroeconomic Impact of Remittances in Ghana.” Bank of
Ghana. Accra.
Addy, David Nii, Boris Wijkstrom, and Colleen Thouez. 2003. “Migrant Remittances – Country
of Origin Experiences.” International Migration Policy Programme. www.impprog.ch. Geneva.
Adepoju, Aderanti. 1988. "International migration in Africa South of the Sahara." In
International Migration Today, Vol.1 Trends and Prospects, edited by Reginald T. Appleyard.
Paris: UNESCO.
_______________. 2002. “Fostering Free Movement of Persons in West Africa: Achievements,
Constraints, and Prospects for Intraregional Migration.” International Migration. 40(2):4-28.
______________. 2005. “Creating a Borderless West Africa: Constraints and Prospects for IntraRegional Migration.” UNESCO
AFFORD. 2000. “Globalisation and Development: A Diaspora Dimension.” African Foundation
for Development. London.
Afolayan, A. A. 1988. “Immigration and Expulsion of ECOWAS Aliens in Nigeria.”
International Migration Review. 22(1):4-27.
___________. 2001. “Issues and Challenges of Emigration Dynamics in Developing Countries.”
International Migration. 39(4):1-38.
Ammassari, Savina. 2005. “Gestion des migrations et politiques de developpement: Optimiser
les benefices de la migration internationale en Afrique de l’Ouest.” International Labour Office.
Geneva.
Anarfi, John and Stephen Kwankye with Ofuso-Mensah Ababio and Richmond Tiemoko. 2003.
“Migration to and from Ghana.” Working Paper C4. Development Research Centre on
Migration, Globalisation and Poverty. University of Sussex.
Antropologi. 2005. “Burkina Faso: Thousands of migrants now living as strangers in their
homeland.” www.antropologi.info
Aredo, Dejene. 2005. Migrant Remittances, Shocks and Poverty in Urban Ethiopia: An Analysis
of Micro Level Panel Data. Addis Ababa University.
50
Asiedu, Alex. 2004. “Some benefits of migrants’ return visits to Ghana.” Population, Space and
Place. 11(1):1-11.
Asmelash, Bruk. 2006. “Overview of Remittance in Ethiopia.” Prepared for the Fourth
International Conference on the Ethiopian Economy. Addis Ababa. April.
Azam, Jean-Paul and Flore Gubert. 2004. “Those in Kayes: The Impact of Remittances on their
Recipients in Africa.” University of Toulouse and DIAL, IRD, Paris.
Ba, Hamidou. 2005. “Legislations Relatives aux Travailleurs Migrants en Afrique de l’Ouest.”
International Labor Office. Dakar.
Bach, Stephen. 2003. “International Migration of Health Workers: Labour and Social Issues.”
Sectoral Activities Programme Working Paper. International Labour Office. Geneva.
Banki, Susan. 2004. “Refugee integration in the intermediate term: a study of Nepal, Pakistan,
and Kenya.” New Issues in Refugee Research Working Paper No. 108. Evaluation and Policy
Analysis Unit. United Nations High Commissioner for Refugees. Geneva.
Bhorat, Haroon, Jean-Baptiste Meyer and Cecil Mlatsheni. 2002. “Skilled Labour Migration
from Developing Countries: Study on South and Southern Africa.” International Migration
Papers No. 52. International Labour Office. Geneva.
Birdsall, Nancy., Behrman,J. & Chuhan, Punam. 1988. "Client choice of health care treatment in
rural Mali.' Paper presented at annual meeting of Population Association of America, New
Orleans, April.
Black, Richard. 2004?. “Migration and pro-poor policies in Africa.”Development Research
Centre on Migration, Globalisation and Poverty.” University of Sussex.
Black, Richard, Russell King, and Richmond Tiemoko. 2003a. “International Workshop on
Migration and Poverty in West Africa.” March 13-14. University of Sussex.
Black, Richard, Russell King, and Julie Lichfield. 2003b. “Transnational Migration, Return and
Development in West Africa.” Centre for Migration Research. University of Sussex.
Black, Richard, Lyndsay McLean Hilker, and Claire Pooley. 2004. “Migration and Pro-Poor
Policy in East Africa.” Development Research Centre on Migration , Globalisation and Poverty.
University of Sussex.
Black, Richard, Claudia Natali and Jessica Skinner. 2005. “Migration and Inequality.” Equity
and Development. Background paper for World Bank’s World Development Report 2006.
Buchan, James, Tina Parkin, and Julie Sochalski. 2003. “International Nurse Mobility: Trends
and Policy Implications.” World Health Organization. WHO/EIP/OSD/2003.3 Geneva.
51
Bump, Micah. 2006. “Ghana: Searching for Opportunities at Home and Abroad.” Migration
Policy Institute. Washington, D.C.
Carling, Jorgen. 2006 “Migration, Human Smuggling and Trafficking from Nigeria to Europe.”
International Organization of Migration.
Castle, Sarah and Aisse Diarra. 2003. “La Migration Internationale des Jeunes Maliens:
Tradition, Necessite, ou Rite de Passage ?. London School of Hygiene and Tropical Medicine.
CGAP 2003 “Supporting remittances in Southern Africa.” Consultative Group to Assist the Poor.
www.cgap.org.
Clemens, Michael and Gunilla Pettersson. 2007. “New Data on African Health Professionals
Abroad.” Center for Global Development Working Paper No. 95.
Cobbe, Jim. 2004. “Lesotho: Will the Enclave Empty?” Migration Information Source.
Migration Policy Institute. Washington, D.C.
Coplan, David B. 2001. “A River Runs Through It: The Meaning of the Lesotho-Free State
Border.” African Affairs. 100:81-116.
Crush, Jonathan and David A. McDonald. 2001. “Introduction to Special Issue: Evaluating South
African Immigration Policy after Apartheid.”Africa Today.
Crush, Jonathan and Sally Peberdy. 2003. “Criminal Tendencies: Immigrants and Illegality in
South Africa.” Migration Policy Brief No.10. Southern African Migration Project. Johannesburg.
Danso, Ransford and David A. McDonald. 2001. “Writing Xenophobia: Immigration and the
Print Media in Post-apartheid South Africa.” Africa Today
Delap, Emily, Boureima Ouedraogo and Bakary Sogoba. 2005. “Developing Alternatives to the
Worst Forms of Child Labour in Mali and Burkina Faso.” Save the Children. United Kingdom.
De Souza, Roger-Mark. 2006. “Using Return Migration As a Development Tool—Are the Right
Policies in Place?” Population Reference Bureau. Washington, D.C.
Diatta, Marie Angelique and Ndiaga Mbow. 1999. “Releasing the Development Potential of
Return Migration: The Case of Senegal.” International Migration. 37(1):243-266.
Docquier, Frederic and Abdeslam Marfouk. 2004. “Measuring the international mobility of
skilled workers (1990-2000): release 1.0.” Policy Research Working Paper Series 3381. World
Bank
EU. 2006. “EU-Mauritius Country Strategy Paper.” European Union. Brussels.
52
Fadayomi, T. O. 1994. “Brain Drain from African States: Empirical Evidence and Policy
Implications.” African Population Studies. 9(1).
Fall, Abdou Salam. 2000 “Enjeux et defies de la migration internationale de travail ouestafricaine.” Cahiers de Migrations Internationales. International Labor Office. Geneva.
Fall, Papa Demba. 2003. “Migration internationale et droits des travailleurs au Senegal.”
UNESCO. SHS/2003/MC/5.
Findley, Sally E. 1989. "Choosing Between African and French Destinations: the Role of Family
and Community Factors in Migration in the Senegal River Valley. Paper presented at
Department de Demographie, Universite de Montreal, March 14.
_____________. 2004. “Mali: Seeking Opportunity Abroad.” Migration Information Source.
Migration Policy Institute. Washington, D.C.
Findley, Sally E. and Salif Sow. 1998. “From Season to Season: Agriculture, Poverty and
Migration in the Senegal River Valley, Mali.” In Appleyard, Reginald, ed. Emigration Dynamics
in Developing Countries; Volume 1: Sub-Saharan Africa. Ashgate:Brookfield,
Vermond/Aldershot, England
Gastrow, Peter. 1999. “Main trends in the development of South Africa's organised crime.”
African Security Review. 8(6)59-65.
Gay, 2000. “Lesotho and South Africa: time for a new immigration compact.” In David A.
McDonald, ed., On Borders: Perspectives on International Migration in Southern Africa. St.
Martin’s Press. New York.
Genesis. 2003. “African Families, African Money: Bridging the Money Transfer Divide.” Study
done for FinMark Trust. Genesis Analytics. Johannesburg. April.
Golooba-Mutebi, Frederick and Stephen M Tollman. 2004. “Survival to livelihood strategies
for Mozambican refugees in South Africa.” Forced Migration Review. 20(28-29).
Gubert, Flore. 2005. “Migrant Remittances and their Impact on Development in the Home
Economies: The Case of Africa.” In Migration, Remittances and Development. OECD. Paris.
Guilmoto, Christophe Z. 1998. “Institutions and Migrations: Short-Term versus Long-Term
Moves in Rural West Africa.” Population Studies. 52:85-103.
Gupta, Sanjeev, Catherine Pattillo, and Smita Wagh. 2006. “Impact of Remittances on Poverty
and Financial Development in Sub-Saharan Africa.” International Monetary Fund Working
Paper. December. Washington D.C.
53
Gustafsson, Bjorn and Negatu Makonnen. 1993. “Poverty and Remittances in Lesotho.” Journal
of African Economies. 2(1):49-73.
Hampshire, K and S. Randall. 1999. “Seasonal labour migration strategies in the Sahel: coping
with poverty or optimizing security?” International Journal of Population Geography. 5(5):36785.
Hearing, Liesbeth and Rob van der Erf. 2001. “Why do people migrate?” Eurostat.
Hennop, Ettienne, Clare Jefferson, and Andrew McLean. 2001. “The Challenge to Control South
Africa’s Borders and Borderline.” Monograph 57. www.issafrica.org.
Human Rights Watch. 2006. “Unprotected Migrants: Zimbabweans in South Africa’s Limpopo
Province.” 18(6). July.
IOM. 2005a. “Policy Approaches to Migration and Development.” MC/INIF/281. International
Organization of Migration.
_____. 2005b. World Migration. International Organization of Migration.
Kabki, Mirjam, Valentina Mazzucato, and Ernest Appiah. 2004. “’Wo benane a eye bebree’: The
economic impact of remittances of Netherlands-based Ghanaian migrants on rural Ashanti.”
Populatoin and Space. 10(2):85-97.
Kebede, Emebet. 2001. “Ethiopia: An Assessment of the International Labour Migration
Situation: The Case of Female Migrants.” GENPROM Working Paper No. 3. Gender Promotion
Program. International Labour Office. Geneva.
Klaaren, Jonathan and Jaya Ramji. 2001. “Inside Illegality: Migration Policing in South Africa
after Apartheid.” Africa Today (stapled to Crush and McDonald)
Konseiga, Adama. 2005a. “New Patterns in the Human Migration in West Africa.”
_____________. 2005b. “Household Migration Decisions as Survival Strategy: The Case of
Burkina Faso.” IZA Discussion Paper No. 1819. Institute for the Study of Labor. Bonn.
____________. 2005c. “Regionalism in West Africa: Do Polar Countries Reap the Benefits? A
Role for Migration.” Institute for the Study of Labor Discussion Paper 1516. Bonn
Lachaud, Jean-Pierre. 1999 “Envoi de fonds, inegalite et pauvrete au Burkina Faso.”
Lassailly-Jacob, Veronique, Florence Boyer, and Julien Brachet. 2006. “South-South Migration:
Example of Sub-Saharan Africa.” European Parliament.
54
Litchfield, Julie and Hugh Waddington. 2003. “Migration and Poverty in Ghana: Evidence from
the Ghana Living Standards Survey.” Sussex Migration Working Paper No. 10. Sussex Centre
for Migration Research.
Loup, Jackques. 2005. “The Economy of Solidarity: Expatriate Workers’ Remittances to SubSaharan Africa.” Foundation pour L’Innovation Politique. June.
Lucas, Robert E.B. 1987. “Emigration to South Africa’s Mines.” The American Economic
Review. 77(3):313-330.
Lucas, Robert E. B. and Oded Stark. 1985.”Motivations to Remit: Evidence from Botswana.”
Economic Journal. 93:901-18.
Lurie, Mark N. 2004. “Migration, Sexuality and the Spread of HIV/AIDS in Rural South
Africa.” Southern African Migration Project.
Macchiavello, Michela. 2003. Forced migrants as an under-utilized asset: refugee skills,
livelihoods, and achievements in Kampala, Uganda. New Issues in Refugee Research. Working
Paper No. 95. United Nations High Commission on Refugees. Geneva.
Madakufamba. 2005. “Cross-border movement of people in southern Africa.” Regional
Economic Development and Integration News Features.” No. 82. October
Maharaj, Brij. 2004. “Immigration to post-apartheid South Africa.” Global Migration
Perspectives No. 1. Global Commission on International Migration. Geneva.
Martens, Jonathan, Maciej ‘Mac’ Pieczkowski, Bernadette van Vuuren-Smyth. 2003.
“Seduction, Sale and Slavery: Trafficking in Women & Children for Sexual Exploitation in
Southern Africa.” International Organization of Migration Regional Office for Southern Africa.
Pretoria.
Martin, Philip, Susan Martin and Patrick Weil. 2002. “Best Practice Options: Mali.”
International Migration. 40(3):87-101.
Mattes, Robert and Wayne Richmond. 2000. “The Brain Drain: What do Skilled South Africans
Think?” in Losing our Minds: Skilled Migration and the South African Brain Drain, Crush,
Jonathan, David McDonald and Vincent Williams, eds. Southern African Migration Project.
Johannesburg.
Mazzucato, Valentina, Bart van den Boom, and N. N. N. Nsowah-Nuamah. 2005?. “The impact
of international remittances on local living standards: Evidence for households in Ghana.”
Mimeo. Ghana Transnational Networks Research Program.
McDonald, David and Jonathan Crush. 2002. “Thinking About the Brain Drain in Southern
Africa.” Southern Africa Migration Project. Migration Policy Brief No. 8.
55
Minnaar, Anthony. 2001. “Border Control and Regionalism: The Case of South Africa.”African
Security Review. 10(2):
Mochebelele, Motsamai T. and Alex Winter-Nelson. 2000. “Migrant Labor and Farm Technical
Efficency in Lesotho.” World Development. 28(1):143-153.
Mutume, Gumisai. 2002. “Reversing Africa’s ‘brain drain’. Africa Recovery. 17(2)1-9.
Myburgh, Andrew. 2004. “Explaining Emigration from South Africa.” 72(1):122-148.
Nunn A. J., H. U. Wagner A. Kamali, J. F. Kengeya-Kayondo, and D. W. Mulder. 1995.
“Migration and HIV-1 seroprevalence in a rural Ugandan population.” AIDS 9(5):503-506.
Nwajiuba, Chinedum. 2005. “International Migration and Livelihoods in Southeastern Nigeria.”
Global Migration Perspectives No. 50. Global Commission on International Migration.
OECD 2003. “The International Mobility of Health Professionals: An Evaluation and Analysis
Based on the Case of South Africa.” Trends in International Migration. Organization for
Economic Cooperation and Development. Paris.
Oucho, John O. 2006. “Migration and refugees in Eastern Africa: A challenge for the East
African Community.” In Views on Migration in Sub-Saharan Africa. Edited by Catherine Cross,
Derik Gelderblom, Niel Roux, and Jonathan Mafukidze. HSRC Press.
Peberdy, Sally and Christian Rogerson. 2000. “Transnationalism and Non-South African
Entrepreneurs in South Africa’s Small, Medium and Micro-Enterprise (SMME) Economy.”
Canadian Journal of African Studies. 34(1):20-40.
Pizarro, Gabriela Rodriguez. 2006. “Specific Groups and Individuals: Migrant Workers.” UN
Economic and Social Council. Commission on Human Rights. E/CN.4/2006/73/Add.2
Plath, Joel C., Holland, D.W. & Carvalho, J.W. 1987. Labor migration in
Southern Africa and agricultural development: some lessons from Lesotho.
Journal of Developing Areas. 21:159-76.
Quartey, Peter. 2006. “Migration and Development: Challenges and Opportunities for Sending
Countries Ghana Country Case Study.” Report prepared for German Marshall Fund of the USA.
Washington, D.C. July 22-26.
Quartey, Peter and Theresa Blankson. 2004. “Do Migrant Remittances Minimize the Impact of
Macro-Volatility on the Poor in Ghana?” Global Development Network.
Rado, Emil. 1986. “Notes towards a political economy of Ghana today.” African Affairs.
85(341):563-572.
Raunet, Mireille. 2005. “Motivating Migrants for Social and Economic Developments in Mali
and Senegal.” In Migration, Remittances and Development. OECD. Paris.
56
Riccio, Bruno. 2003. “More than a Trade Diaspora: Senegalese Transnational Experiences in
Emilia-Romagna.” In Khalid Koser, ed. New African Diasporas. Routledge : London.
Russell, Stanton Sharon, Karen Jacobsen, and William Deane Stanley. 1990. “International
Migration and Development in Sub-Saharan Africa”. World Bank Discussion Papers Africa
Technical Department Series No. 101. Washington D.C.
SADC Today. 2007. “Visa-free entry for key tourist source countries by 2008.”
[email protected]
Sall, Babacar. 2005. “Migration, Remittances and Economic Initiatives in Sub-Saharan Africa.”
In Migration, Remittances and Development. OECD. Paris.
Sander, Crestin and Issa Barro. 2003. “Etude sur le Transfert d’Argent des Emigres au Senegal et
les Services de Transfert en Microfinance.” Working Paper No. 40. Social Finance Program.
International Labor Office. Geneva.
Sander, Cerstin and Samuel Munzele Maimbo. 2003. “Migrant Labor Remittances in Africa:
Reducing Obstacles to Development Contributions.” Africa Region Working Paper Series No.
64. World Bank. Washington, D.C.
Sander, Cerstin, Peter Mukwana and Altemius Millinga. ?. “Passing the Buck. Money Transfer
Systems: The Practice and Potential for Products in Tanzania and Uganda.” MicroSave-Africa &
Australian Development Cooperation.
Shaw, Mark. 2001. “Towards an understanding of West African criminal networks in Southern
Africa.” Africa Security Review. 10(4)
Solomon, Hussein. 1996. “Strategic Perspectives on Illegal Immigration into South Africa.”
African Security Review. 5(4)
Sparreboom, Theo and Pete Sparreboom-Buger. 1996. “Migrant worker remittances in Lesotho:
A Review of the Deferred Pay Scheme.” Working Paper No. 16. Enterprise and Cooperative
Development Department. International Labour Office. Geneva.
Tall, Serigne Mansour. 2003. “Les émigrés senegalais et les nouvelles technologies de
l’information et de la communication.” Technologie, enterprise et societe, document du
programme No. 7. Institut de recherché des Nations Unies pour le developpement social.
Thiam, Macoumba, Rebecca Pery, and Victor Piche. 2004. “Migration and HIV in Northern
Senegal.” Population Reference Bureau. www.prb.org.
Thiombiano, Taladidia. 2000. “L’Exode des Competences au Burkina Faso.” Centre d’Etudes,
des Documentation, de Recherches Economomiques et Sociales. Universite de Ouagadougoug.
Burkina Faso.
57
Tiemoko, Richard. (2004) “Understanding the Opportunities for and Challenges to Migrants’
Investment in Côte d’Ivoire and Ghana.” In At home in the world? : international migration and
development in contemporary Ghana and West Africa . edited by Takyiwaa Manuh
Tsegai, Daniel and Maria Plotnikova. 2004. “Migration and Household Income Differentials in
the Volta Basin of Ghana: A Sample Selection Approach.” Paper presented at Canadian
Economics Association 38th Annual Meetings. June.
UN. 2004. World Economic and Social Survey 2004. United Nations Department of Economic
and Social Affairs. New York.
UNCTAD. 2006. “Using Commoditized Revenue Flows to Leverage Access to International
Finance: With a Special Focus on Migrant Remittances and Payment Flows.” United Nations
Conference on Trade and Development. UNCTAD/DITC/COM/2006/9
Van Rooyen, Johann. 2003. The New Great Trek: The Story of South Africa’s White Exodus.
University of South Africa.
Van Vlaenderen, Hilde, Serigne Mansour Tall, and Gora Gaye. 2004. “Senegal” in Cotula,
Lorenzo and Camilla Toulmin, eds. Till to tiller: International migration, remittances and land
rights in West Africa. International Institute for Environment and Development. Issue Paper no.
132. London.
Wocke, Albert and Saul Klein. 2002. “The implications of South Africa’s skills migration policy
for country competitiveness.” Development Southern Africa. 19(4):441-454.
World Bank. 1993. Mali: Assessment of Living Conditions.
World Bank. 2006. Global Economic Prospects 2006.
Wouterse, Fleur and Marrit van den Berg. 2004. “Migration for Survival or Accumulation:
Evidence from Burkina Faso.” Wageningen University and Research Center, Mansholt Graduate
School of Social Sciences.
Yameogo, K Robert, Robert T. Perry, Andre Yameogo, Chantal Kambire, M. Kader Konde,
Deogratias Nshimirimana, Robert Kezaala, Bradley S. Hersh, K. Lisa Cairns, and ZPeter Strebel.
2005. “Migration as a risk factor for measles after a mass vaccination campaign.” International
Journal of Epidemiology. 34(3):556-564.
58